Статті в журналах з теми "Task-specific representation learnining"
Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Task-specific representation learnining.
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 статей у журналах для дослідження на тему "Task-specific representation learnining".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте статті в журналах для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Wang, Chaoqun, Xuejin Chen, Shaobo Min, Xiaoyan Sun, and Houqiang Li. "Task-Independent Knowledge Makes for Transferable Representations for Generalized Zero-Shot Learning." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 3 (May 18, 2021): 2710–18. http://dx.doi.org/10.1609/aaai.v35i3.16375.
Повний текст джерелаАнотація:
Generalized Zero-Shot Learning (GZSL) targets recognizing new categories by learning transferable image representations. Existing methods find that, by aligning image representations with corresponding semantic labels, the semantic-aligned representations can be transferred to unseen categories. However, supervised by only seen category labels, the learned semantic knowledge is highly task-specific, which makes image representations biased towards seen categories. In this paper, we propose a novel Dual-Contrastive Embedding Network (DCEN) that simultaneously learns task-specific and task-independent knowledge via semantic alignment and instance discrimination. First, DCEN leverages task labels to cluster representations of the same semantic category by cross-modal contrastive learning and exploring semantic-visual complementarity. Besides task-specific knowledge, DCEN then introduces task-independent knowledge by attracting representations of different views of the same image and repelling representations of different images. Compared to high-level seen category supervision, this instance discrimination supervision encourages DCEN to capture low-level visual knowledge, which is less biased toward seen categories and alleviates the representation bias. Consequently, the task-specific and task-independent knowledge jointly make for transferable representations of DCEN, which obtains averaged 4.1% improvement on four public benchmarks.
2
Li, Yingcong, and Samet Oymak. "Provable Pathways: Learning Multiple Tasks over Multiple Paths." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 7 (June 26, 2023): 8701–10. http://dx.doi.org/10.1609/aaai.v37i7.26047.
Повний текст джерелаАнотація:
Constructing useful representations across a large number of tasks is a key requirement for sample-efficient intelligent systems. A traditional idea in multitask learning (MTL) is building a shared representation across tasks which can then be adapted to new tasks by tuning last layers. A desirable refinement of using a shared one-fits-all representation is to construct task-specific representations. To this end, recent PathNet/muNet architectures represent individual tasks as pathways within a larger supernet. The subnetworks induced by pathways can be viewed as task-specific representations that are composition of modules within supernet's computation graph. This work explores the pathways proposal from the lens of statistical learning: We first develop novel generalization bounds for empirical risk minimization problems learning multiple tasks over multiple paths (Multipath MTL). In conjunction, we formalize the benefits of resulting multipath representation when adapting to new downstream tasks. Our bounds are expressed in terms of Gaussian complexity, lead to tangible guarantees for the class of linear representations, and provide novel insights into the quality and benefits of a multipath representation. When computation graph is a tree, Multipath MTL hierarchically clusters the tasks and builds cluster-specific representations. We provide further discussion and experiments for hierarchical MTL and rigorously identify the conditions under which Multipath MTL is provably superior to traditional MTL approaches with shallow supernets.
3
Wang, Gerui, and Sheng Tang. "Generalized Zero-Shot Image Classification via Partially-Shared Multi-Task Representation Learning." Electronics 12, no. 9 (May 3, 2023): 2085. http://dx.doi.org/10.3390/electronics12092085.
Повний текст джерелаАнотація:
Generalized Zero-Shot Learning (GZSL) holds significant research importance as it enables the classification of samples from both seen and unseen classes. A prevailing approach for GZSL is learning transferable representations that can generalize well to both seen and unseen classes during testing. This approach encompasses two key concepts: discriminative representations and semantic-relevant representations. “Semantic-relevant” facilitates the transfer of semantic knowledge using pre-defined semantic descriptors, while “discriminative” is crucial for accurate category discrimination. However, these two concepts are arguably inherently conflicting, as semantic descriptors are not specifically designed for image classification. Existing methods often struggle with balancing these two aspects and neglect the conflict between them, leading to suboptimal representation generalization and transferability to unseen classes. To address this issue, we propose a novel partially-shared multi-task representation learning method, termed PS-GZSL, which jointly preserves complementary and sharable knowledge between these two concepts. Specifically, we first propose a novel perspective that treats the learning of discriminative and semantic-relevant representations as optimizing a discrimination task and a visual-semantic alignment task, respectively. Then, to learn more complete and generalizable representations, PS-GZSL explicitly factorizes visual features into task-shared and task-specific representations and introduces two advanced tasks: an instance-level contrastive discrimination task and a relation-based visual-semantic alignment task. Furthermore, PS-GZSL employs Mixture-of-Experts (MoE) with a dropout mechanism to prevent representation degeneration and integrates a conditional GAN (cGAN) to synthesize unseen features for estimating unseen visual features. Extensive experiments and more competitive results on five widely-used GZSL benchmark datasets validate the effectiveness of our PS-GZSL.
4
Basu Roy Chowdhury, Somnath, and Snigdha Chaturvedi. "Sustaining Fairness via Incremental Learning." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 6 (June 26, 2023): 6797–805. http://dx.doi.org/10.1609/aaai.v37i6.25833.
Повний текст джерелаАнотація:
Machine learning systems are often deployed for making critical decisions like credit lending, hiring, etc. While making decisions, such systems often encode the user's demographic information (like gender, age) in their intermediate representations. This can lead to decisions that are biased towards specific demographics. Prior work has focused on debiasing intermediate representations to ensure fair decisions. However, these approaches fail to remain fair with changes in the task or demographic distribution. To ensure fairness in the wild, it is important for a system to adapt to such changes as it accesses new data in an incremental fashion. In this work, we propose to address this issue by introducing the problem of learning fair representations in an incremental learning setting. To this end, we present Fairness-aware Incremental Representation Learning (FaIRL), a representation learning system that can sustain fairness while incrementally learning new tasks. FaIRL is able to achieve fairness and learn new tasks by controlling the rate-distortion function of the learned representations. Our empirical evaluations show that FaIRL is able to make fair decisions while achieving high performance on the target task, outperforming several baselines.
5
Davvetas, Athanasios, Iraklis A. Klampanos, Spiros Skiadopoulos, and Vangelis Karkaletsis. "Evidence Transfer: Learning Improved Representations According to External Heterogeneous Task Outcomes." ACM Transactions on Knowledge Discovery from Data 16, no. 5 (October 31, 2022): 1–22. http://dx.doi.org/10.1145/3502732.
Повний текст джерелаАнотація:
Unsupervised representation learning tends to produce generic and reusable latent representations. However, these representations can often miss high-level features or semantic information, since they only observe the implicit properties of the dataset. On the other hand, supervised learning frameworks learn task-oriented latent representations that may not generalise in other tasks or domains. In this article, we introduce evidence transfer, a deep learning method that incorporates the outcomes of external tasks in the unsupervised learning process of an autoencoder. External task outcomes also referred to as categorical evidence, are represented by categorical variables, and are either directly or indirectly related to the primary dataset—in the most straightforward case they are the outcome of another task on the same dataset. Evidence transfer allows the manipulation of generic latent representations in order to include domain or task-specific knowledge that will aid their effectiveness in downstream tasks. Evidence transfer is robust against evidence of low quality and effective when introduced with related, corresponding, or meaningful evidence.
6
Konidaris, George, Leslie Pack Kaelbling, and Tomas Lozano-Perez. "From Skills to Symbols: Learning Symbolic Representations for Abstract High-Level Planning." Journal of Artificial Intelligence Research 61 (January 31, 2018): 215–89. http://dx.doi.org/10.1613/jair.5575.
Повний текст джерелаАнотація:
We consider the problem of constructing abstract representations for planning in high-dimensional, continuous environments. We assume an agent equipped with a collection of high-level actions, and construct representations provably capable of evaluating plans composed of sequences of those actions. We first consider the deterministic planning case, and show that the relevant computation involves set operations performed over sets of states. We define the specific collection of sets that is necessary and sufficient for planning, and use them to construct a grounded abstract symbolic representation that is provably suitable for deterministic planning. The resulting representation can be expressed in PDDL, a canonical high-level planning domain language; we construct such a representation for the Playroom domain and solve it in milliseconds using an off-the-shelf planner. We then consider probabilistic planning, which we show requires generalizing from sets of states to distributions over states. We identify the specific distributions required for planning, and use them to construct a grounded abstract symbolic representation that correctly estimates the expected reward and probability of success of any plan. In addition, we show that learning the relevant probability distributions corresponds to specific instances of probabilistic density estimation and probabilistic classification. We construct an agent that autonomously learns the correct abstract representation of a computer game domain, and rapidly solves it. Finally, we apply these techniques to create a physical robot system that autonomously learns its own symbolic representation of a mobile manipulation task directly from sensorimotor data---point clouds, map locations, and joint angles---and then plans using that representation. Together, these results establish a principled link between high-level actions and abstract representations, a concrete theoretical foundation for constructing abstract representations with provable properties, and a practical mechanism for autonomously learning abstract high-level representations.
7
Gu, Jie, Feng Wang, Qinghui Sun, Zhiquan Ye, Xiaoxiao Xu, Jingmin Chen, and Jun Zhang. "Exploiting Behavioral Consistence for Universal User Representation." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 5 (May 18, 2021): 4063–71. http://dx.doi.org/10.1609/aaai.v35i5.16527.
Повний текст джерелаАнотація:
User modeling is critical for developing personalized services in industry. A common way for user modeling is to learn user representations that can be distinguished by their interests or preferences. In this work, we focus on developing universal user representation model. The obtained universal representations are expected to contain rich information, and be applicable to various downstream applications without further modifications (e.g., user preference prediction and user profiling). Accordingly, we can be free from the heavy work of training task-specific models for every downstream task as in previous works. In specific, we propose Self-supervised User Modeling Network (SUMN) to encode behavior data into the universal representation. It includes two key components. The first one is a new learning objective, which guides the model to fully identify and preserve valuable user information under a self-supervised learning framework. The other one is a multi-hop aggregation layer, which benefits the model capacity in aggregating diverse behaviors. Extensive experiments on benchmark datasets show that our approach can outperform state-of-the-art unsupervised representation methods, and even compete with supervised ones.
8
Yu, Wenmeng, Hua Xu, Ziqi Yuan, and Jiele Wu. "Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysis." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 12 (May 18, 2021): 10790–97. http://dx.doi.org/10.1609/aaai.v35i12.17289.
Повний текст джерелаАнотація:
Representation Learning is a significant and challenging task in multimodal learning. Effective modality representations should contain two parts of characteristics: the consistency and the difference. Due to the unified multimodal annota- tion, existing methods are restricted in capturing differenti- ated information. However, additional unimodal annotations are high time- and labor-cost. In this paper, we design a la- bel generation module based on the self-supervised learning strategy to acquire independent unimodal supervisions. Then, joint training the multimodal and uni-modal tasks to learn the consistency and difference, respectively. Moreover, dur- ing the training stage, we design a weight-adjustment strat- egy to balance the learning progress among different sub- tasks. That is to guide the subtasks to focus on samples with the larger difference between modality supervisions. Last, we conduct extensive experiments on three public multimodal baseline datasets. The experimental results validate the re- liability and stability of auto-generated unimodal supervi- sions. On MOSI and MOSEI datasets, our method surpasses the current state-of-the-art methods. On the SIMS dataset, our method achieves comparable performance than human- annotated unimodal labels. The full codes are available at https://github.com/thuiar/Self-MM.
9
Ahmed, Mahtab, and Robert E. Mercer. "Modelling Sentence Pairs via Reinforcement Learning: An Actor-Critic Approach to Learn the Irrelevant Words." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 05 (April 3, 2020): 7358–66. http://dx.doi.org/10.1609/aaai.v34i05.6230.
Повний текст джерелаАнотація:
Learning sentence representation is a fundamental task in Natural Language Processing. Most of the existing sentence pair modelling architectures focus only on extracting and using the rich sentence pair features. The drawback of utilizing all of these features makes the learning process much harder. In this study, we propose a reinforcement learning (RL) method to learn a sentence pair representation when performing tasks like semantic similarity, paraphrase identification, and question-answer pair modelling. We formulate this learning problem as a sequential decision making task where the decision made in the current state will have a strong impact on the following decisions. We address this decision making with a policy gradient RL method which chooses the irrelevant words to delete by looking at the sub-optimal representation of the sentences being compared. With this policy, extensive experiments show that our model achieves on par performance when learning task-specific representations of sentence pairs without needing any further knowledge like parse trees. We suggest that the simplicity of each task inference provided by our RL model makes it easier to explain.
10
Mehmood, Tahir, Ivan Serina, Alberto Lavelli, Luca Putelli, and Alfonso Gerevini. "On the Use of Knowledge Transfer Techniques for Biomedical Named Entity Recognition." Future Internet 15, no. 2 (February 17, 2023): 79. http://dx.doi.org/10.3390/fi15020079.
Повний текст джерелаАнотація:
Biomedical named entity recognition (BioNER) is a preliminary task for many other tasks, e.g., relation extraction and semantic search. Extracting the text of interest from biomedical documents becomes more demanding as the availability of online data is increasing. Deep learning models have been adopted for biomedical named entity recognition (BioNER) as deep learning has been found very successful in many other tasks. Nevertheless, the complex structure of biomedical text data is still a challenging aspect for deep learning models. Limited annotated biomedical text data make it more difficult to train deep learning models with millions of trainable parameters. The single-task model, which focuses on learning a specific task, has issues in learning complex feature representations from a limited quantity of annotated data. Moreover, manually constructing annotated data is a time-consuming job. It is, therefore, vital to exploit other efficient ways to train deep learning models on the available annotated data. This work enhances the performance of the BioNER task by taking advantage of various knowledge transfer techniques: multitask learning and transfer learning. This work presents two multitask models (MTMs), which learn shared features and task-specific features by implementing the shared and task-specific layers. In addition, the presented trained MTM is also fine-tuned for each specific dataset to tailor it from a general features representation to a specialized features representation. The presented empirical results and statistical analysis from this work illustrate that the proposed techniques enhance significantly the performance of the corresponding single-task model (STM).
11
Jeannerod, M. "The representing brain: Neural correlates of motor intention and imagery." Behavioral and Brain Sciences 17, no. 2 (June 1994): 187–202. http://dx.doi.org/10.1017/s0140525x00034026.
Повний текст джерелаАнотація:
AbstractThis paper concerns how motor actions are neurally represented and coded. Action planning and motor preparation can be studied using a specific type of representational activity, motor imagery. A close functional equivalence between motor imagery and motor preparation is suggested by the positive effects of imagining movements on motor learning, the similarity between the neural structures involved, and the similar physiological correlates observed in both imaging and preparing. The content of motor representations can be inferred from motor images at a macroscopic level, based on global aspects of the action (the duration and amount of effort involved) and the motor rules and constraints which predict the spatial path and kinematics of movements. A more microscopic neural account calls for a representation of object-oriented action. Object attributes are processed in different neural pathways depending on the kind of task the subject is performing. During object-oriented action, a pragmatic representation is activated in which object affordances are transformed into specific motor schemas (independently of other tasks such as object recognition). Animal as well as human clinical data implicate the posterior parietal and premotor cortical areas in schema instantiation. A mechanism is proposed that is able to encode the desired goal of the action and is applicable to different levels of representational organization.
12
Chen, Yuhao, Alexander Wong, Yuan Fang, Yifan Wu, and Linlin Xu. "Deep Residual Transform for Multi-scale Image Decomposition." Journal of Computational Vision and Imaging Systems 6, no. 1 (January 15, 2021): 1–5. http://dx.doi.org/10.15353/jcvis.v6i1.3537.
Повний текст джерелаАнотація:
Multi-scale image decomposition (MID) is a fundamental task in computer vision and image processing that involves the transformation of an image into a hierarchical representation comprising of different levels of visual granularity from coarse structures to fine details. A well-engineered MID disentangles the image signal into meaningful components which can be used in a variety of applications such as image denoising, image compression, and object classification. Traditional MID approaches such as wavelet transforms tackle the problem through carefully designed basis functions under rigid decomposition structure assumptions. However, as the information distribution varies from one type of image content to another, rigid decomposition assumptions lead to inefficiently representation, i.e., some scales can contain little to no information. To address this issue, we present Deep Residual Transform (DRT), a data-driven MID strategy where the input signal is transformed into a hierarchy of non-linear representations at different scales, with each representation being independently learned as the representational residual of previous scales at a user-controlled detail level. As such, the proposed DRT progressively disentangles scale information from the original signal by sequentially learning residual representations. The decomposition flexibility of this approach allows for highly tailored representations cater to specific types of image content, and results in greater representational efficiency and compactness. In this study, we realize the proposed transform by leveraging a hierarchy of sequentially trained autoencoders. To explore the efficacy of the proposed DRT, we leverage two datasets comprising of very different types of image content: 1) CelebFaces and 2) Cityscapes. Experimental results show that the proposed DRT achieved highly efficient information decomposition on both datasets amid their very different visual granularity characteristics.
13
Lu, Su, Han-Jia Ye, and De-Chuan Zhan. "Tailoring Embedding Function to Heterogeneous Few-Shot Tasks by Global and Local Feature Adaptors." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 10 (May 18, 2021): 8776–83. http://dx.doi.org/10.1609/aaai.v35i10.17063.
Повний текст джерелаАнотація:
Few-Shot Learning (FSL) is essential for visual recognition. Many methods tackle this challenging problem via learning an embedding function from seen classes and transfer it to unseen classes with a few labeled instances. Researchers recently found it beneficial to incorporate task-specific feature adaptation into FSL models, which produces the most representative features for each task. However, these methods ignore the diversity of classes and apply a global transformation to the task. In this paper, we propose Global and Local Feature Adaptor (GLoFA), a unifying framework that tailors the instance representation to specific tasks by global and local feature adaptors. We claim that class-specific local transformation helps to improve the representation ability of feature adaptor. Global masks tend to capture sketchy patterns, while local masks focus on detailed characteristics. A strategy to measure the relationship between instances adaptively based on the characteristics of both tasks and classes endow GLoFA with the ability to handle mix-grained tasks. GLoFA outperforms other methods on a heterogeneous task distribution and achieves competitive results on benchmark datasets.
14
Zhang, Yu, and Dit-Yan Yeung. "Multi-Task Learning in Heterogeneous Feature Spaces." Proceedings of the AAAI Conference on Artificial Intelligence 25, no. 1 (August 4, 2011): 574–79. http://dx.doi.org/10.1609/aaai.v25i1.7909.
Повний текст джерелаАнотація:
Multi-task learning aims at improving the generalization performance of a learning task with the help of some other related tasks. Although many multi-task learning methods have been proposed, they are all based on the assumption that all tasks share the same data representation. This assumption is too restrictive for general applications. In this paper, we propose a multi-task extension of linear discriminant analysis (LDA), called multi-task discriminant analysis (MTDA), which can deal with learning tasks with different data representations. For each task, MTDA learns a separate transformation which consists of two parts, one specific to the task and one common to all tasks. A by-product of MTDA is that it can alleviate the labeled data deficiency problem of LDA. Moreover, unlike many existing multi-task learning methods, MTDA can handle binary and multi-class problems for each task in a generic way. Experimental results on face recognition show that MTDA consistently outperforms related methods.
15
Xu, Huatao, Pengfei Zhou, Rui Tan, Mo Li, and Guobin Shen. "LIMU-BERT." GetMobile: Mobile Computing and Communications 26, no. 3 (October 7, 2022): 39–42. http://dx.doi.org/10.1145/3568113.3568124.
Повний текст джерелаАнотація:
Deep learning greatly empowers Inertial Measurement Unit (IMU) sensors for a wide range of sensing applications. Most existing works require substantial amounts of wellcurated labeled data to train IMU-based sensing models, which incurs high annotation and training costs. Compared with labeled data, unlabeled IMU data are abundant and easily accessible. This article presents a novel representation learning model that can make use of unlabeled IMU data and extract generalized rather than task-specific features. With the representations learned via our model, task-specific models trained with limited labeled samples can achieve superior performances in typical IMU sensing applications, such as Human Activity Recognition (HAR).
16
Jha, Ritesh, Vandana Bhattacharjee, and Abhijit Mustafi. "Transfer Learning with Feature Extraction Modules for Improved Classifier Performance on Medical Image Data." Scientific Programming 2022 (August 23, 2022): 1–10. http://dx.doi.org/10.1155/2022/4983174.
Повний текст джерелаАнотація:
Transfer learning attempts to use the knowledge learned from one task and apply it to improve the learning of a separate but similar task. This article proposes to evaluate this technique’s effectiveness in classifying images from the medical domain. The article presents a model TrFEMNet (Transfer Learning with Feature Extraction Modules Network), for classifying medical images. Feature representations from General Feature Extraction Module (GFEM) and Specific Feature Extraction Module (SFEM) are input to a projection head and the classification module to learn the target data. The aim is to extract representations at different levels of hierarchy and use them for the final representation learning. To compare with TrFEMNet, we have trained three other models with transfer learning. Experiments on the COVID-19 dataset, brain MRI binary classification, and brain MRI multiclass data show that TrFEMNet performs comparably to the other models. Pretrained model ResNet50 trained on a large image dataset, the ImageNet, is used as the base model.
17
Anam, Mamoona, Dr Kantilal P. Rane, Ali Alenezi, Ruby Mishra, Dr Swaminathan Ramamurthy, and Ferdin Joe John Joseph. "Content Classification Tasks with Data Preprocessing Manifestations." Webology 19, no. 1 (January 20, 2022): 1413–30. http://dx.doi.org/10.14704/web/v19i1/web19094.
Повний текст джерелаАнотація:
Deep reinforcement learning has a major hurdle in terms of data efficiency. We solve this challenge by pretraining an encoder with unlabeled input, which is subsequently finetuned on a tiny quantity of task-specific input. We use a mixture of latent dynamics modelling and unsupervised goal-conditioned RL to encourage learning representations that capture various elements of the underlying MDP. Our approach significantly outperforms previous work combining offline representation pretraining with task-specific finetuning when limited to 100k steps of interaction on Atari games (equivalent to two hours of human experience) and compares favourably with other pretraining methods that require orders of magnitude more data. When paired with larger models and more diverse, task-aligned observational data, our methodology shows great promise, nearing human-level performance and data efficiency on Atari in the best-case scenario.
18
Yuan, Zixuan, Hao Liu, Renjun Hu, Denghui Zhang, and Hui Xiong. "Self-Supervised Prototype Representation Learning for Event-Based Corporate Profiling." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 5 (May 18, 2021): 4644–52. http://dx.doi.org/10.1609/aaai.v35i5.16594.
Повний текст джерелаАнотація:
Event-based corporate profiling aims to assess the evolving operational status of the corresponding corporate from its event sequence. Existing studies on corporate profiling have partially addressed the problem via (i) case-by-case empirical analysis by leveraging traditional financial methods, or (ii) the automatic profile inference by reformulating the problem into a supervised learning task. However, both approaches heavily rely on domain knowledge and are labor-intensive. More importantly, the task-specific nature of both approaches prevents the obtained corporate profiles from being applied to diversified downstream applications. To this end, in this paper, we propose a Self-Supervised Prototype Representation Learning (SePaL) framework for dynamic corporate profiling. By exploiting the topological information of an event graph and exploring self-supervised learning techniques, SePaL can obtain unified corporate representations that are robust to event noises and can be easily fine-tuned to benefit various down-stream applications with only a few annotated data. Specifically, we first infer the initial cluster distribution of noise-resistant event prototypes based on latent representations of events. Then, we construct four permutation-invariant self-supervision signals to guide the representation learning of the event prototype. In terms of applications, we exploit the learned time-evolving corporate representations for both stock price spike prediction and corporate default risk evaluation. Experimental results on two real-world corporate event datasets demonstrate the effectiveness of SePaL for these two applications.
19
Heyes, C. M., and C. L. Foster. "Motor learning by observation: Evidence from a serial reaction time task." Quarterly Journal of Experimental Psychology Section A 55, no. 2 (April 2002): 593–607. http://dx.doi.org/10.1080/02724980143000389.
Повний текст джерелаАнотація:
This study sought evidence of observational motor learning, a type of learning in which observation of the skilled performance of another person not only facilitates motor skill acquisition but does so by contributing to the formation of effector-specific motor representations. Previous research has indicated that observation of skilled performance engages cognitive processes similar to those occurring during action execution or physical practice, but has not demonstrated that these include processes involved in effector-specific representation. In two experiments, observer subjects watched the experimenter performing a serial reaction time (SRT) task with a six-item unique sequence before sequence knowledge was assessed by response time and/or free generation measures. The results suggest that: (1) subjects can acquire sequence information by watching another person performing the task (Experiments 1-2); (2) observation results in as much sequence learning as task practice when learning is measured by reaction times (RTs) and more than task practice when sequence learning is measured by free generation performance (Experiment 2, Part 1); and (3) sequence knowledge acquired by model observation can be encoded motorically—that is, in an effector-specific fashion (Experiment 2, Part 2).
20
Nishida, Satoshi, Yusuke Nakano, Antoine Blanc, Naoya Maeda, Masataka Kado, and Shinji Nishimoto. "Brain-Mediated Transfer Learning of Convolutional Neural Networks." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 04 (April 3, 2020): 5281–88. http://dx.doi.org/10.1609/aaai.v34i04.5974.
Повний текст джерелаАнотація:
The human brain can effectively learn a new task from a small number of samples, which indicates that the brain can transfer its prior knowledge to solve tasks in different domains. This function is analogous to transfer learning (TL) in the field of machine learning. TL uses a well-trained feature space in a specific task domain to improve performance in new tasks with insufficient training data. TL with rich feature representations, such as features of convolutional neural networks (CNNs), shows high generalization ability across different task domains. However, such TL is still insufficient in making machine learning attain generalization ability comparable to that of the human brain. To examine if the internal representation of the brain could be used to achieve more efficient TL, we introduce a method for TL mediated by human brains. Our method transforms feature representations of audiovisual inputs in CNNs into those in activation patterns of individual brains via their association learned ahead using measured brain responses. Then, to estimate labels reflecting human cognition and behavior induced by the audiovisual inputs, the transformed representations are used for TL. We demonstrate that our brain-mediated TL (BTL) shows higher performance in the label estimation than the standard TL. In addition, we illustrate that the estimations mediated by different brains vary from brain to brain, and the variability reflects the individual variability in perception. Thus, our BTL provides a framework to improve the generalization ability of machine-learning feature representations and enable machine learning to estimate human-like cognition and behavior, including individual variability.
21
RAJENDRAN, SRIVIDHYA, and MANFRED HUBER. "LEARNING TASK-SPECIFIC SENSING, CONTROL AND MEMORY POLICIES." International Journal on Artificial Intelligence Tools 14, no. 01n02 (February 2005): 303–27. http://dx.doi.org/10.1142/s0218213005002119.
Повний текст джерелаАнотація:
AI agents and robots that can adapt and handle multiple tasks in real time promise to be a powerful tool. To address the control challenges involved in such systems, the underlying control approach has to take into account the important sensory information. Modern sensors, however, can generate huge amounts of data, rendering the processing and representation of all sensor data in real time computationally intractable. This issue can be addressed by developing task-specific focus of attention strategies that limit the sensory data that is processed at any point in time to the data relevant for the given task. Alone, however, this mechanism is not adequate for solving complex tasks since the robot also has to maintain selected pieces of past information. This necessitates AI agents and robots to have the capability to remember significant past events that are required for task completion. This paper presents an approach that considers focus of attention as a problem of selecting controller and feature pairs to be processed at any given point in time to optimize system performance. This approach is further extended by incorporating short term memory and a learned memory management policy. The result is a system that learns control, sensing, and memory policies that are task-specific and adaptable to real world situations using feedback from the world in a reinforcement learning framework. The approach is illustrated using table cleaning, sorting, stacking, and copying tasks in the blocks world domain.
22
Sun, Kai, Richong Zhang, Samuel Mensah, Yongyi Mao, and Xudong Liu. "Progressive Multi-task Learning with Controlled Information Flow for Joint Entity and Relation Extraction." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 15 (May 18, 2021): 13851–59. http://dx.doi.org/10.1609/aaai.v35i15.17632.
Повний текст джерелаАнотація:
Multitask learning has shown promising performance in learning multiple related tasks simultaneously, and variants of model architectures have been proposed, especially for supervised classification problems. One goal of multitask learning is to extract a good representation that sufficiently captures the relevant part of the input about the output for each learning task. To achieve this objective, in this paper we design a multitask learning architecture based on the observation that correlations exist between outputs of some related tasks (e.g. entity recognition and relation extraction tasks), and they reflect the relevant features that need to be extracted from the input. As outputs are unobserved, our proposed model exploits task predictions in lower layers of the neural model, also referred to as early predictions in this work. But we control the injection of early predictions to ensure that we extract good task-specific representations for classification. We refer to this model as a Progressive Multitask learning model with Explicit Interactions (PMEI). Extensive experiments on multiple benchmark datasets produce state-of-the-art results on the joint entity and relation extraction task.
23
Zhao, Jiabao, Yifan Yang, Xin Lin, Jing Yang, and Liang He. "Looking Wider for Better Adaptive Representation in Few-Shot Learning." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 12 (May 18, 2021): 10981–89. http://dx.doi.org/10.1609/aaai.v35i12.17311.
Повний текст джерелаАнотація:
Building a good feature space is essential for the metric-based few-shot algorithms to recognize a novel class with only a few samples. The feature space is often built by Convolutional Neural Networks (CNNs). However, CNNs primarily focus on local information with the limited receptive field, and the global information generated by distant pixels is not well used. Meanwhile, having a global understanding of the current task and focusing on distinct regions of the same sample for different queries are important for the few-shot classification. To tackle these problems, we propose the Cross Non-Local Neural Network (CNL) for capturing the long-range dependency of the samples and the current task. CNL extracts the task-specific and context-aware features dynamically by strengthening the features of the sample at a position via aggregating information from all positions of itself and the current task. To reduce losing important information, we maximize the mutual information between the original and refined features as a constraint. Moreover, we add a task-specific scaling to deal with multi-scale and task-specific features extracted by CNL. We conduct extensive experiments for validating our proposed algorithm, which achieves new state-of-the-art performances on two public benchmarks.
24
Kumar, Sajit, Alicia Nanelia, Ragunathan Mariappan, Adithya Rajagopal, and Vaibhav Rajan. "Patient Representation Learning From Heterogeneous Data Sources and Knowledge Graphs Using Deep Collective Matrix Factorization: Evaluation Study." JMIR Medical Informatics 10, no. 1 (January 20, 2022): e28842. http://dx.doi.org/10.2196/28842.
Повний текст джерелаАнотація:
Background Patient representation learning aims to learn features, also called representations, from input sources automatically, often in an unsupervised manner, for use in predictive models. This obviates the need for cumbersome, time- and resource-intensive manual feature engineering, especially from unstructured data such as text, images, or graphs. Most previous techniques have used neural network–based autoencoders to learn patient representations, primarily from clinical notes in electronic medical records (EMRs). Knowledge graphs (KGs), with clinical entities as nodes and their relations as edges, can be extracted automatically from biomedical literature and provide complementary information to EMR data that have been found to provide valuable predictive signals. Objective This study aims to evaluate the efficacy of collective matrix factorization (CMF), both the classical variant and a recent neural architecture called deep CMF (DCMF), in integrating heterogeneous data sources from EMR and KG to obtain patient representations for clinical decision support tasks. Methods Using a recent formulation for obtaining graph representations through matrix factorization within the context of CMF, we infused auxiliary information during patient representation learning. We also extended the DCMF architecture to create a task-specific end-to-end model that learns to simultaneously find effective patient representations and predictions. We compared the efficacy of such a model to that of first learning unsupervised representations and then independently learning a predictive model. We evaluated patient representation learning using CMF-based methods and autoencoders for 2 clinical decision support tasks on a large EMR data set. Results Our experiments show that DCMF provides a seamless way for integrating multiple sources of data to obtain patient representations, both in unsupervised and supervised settings. Its performance in single-source settings is comparable with that of previous autoencoder-based representation learning methods. When DCMF is used to obtain representations from a combination of EMR and KG, where most previous autoencoder-based methods cannot be used directly, its performance is superior to that of previous nonneural methods for CMF. Infusing information from KGs into patient representations using DCMF was found to improve downstream predictive performance. Conclusions Our experiments indicate that DCMF is a versatile model that can be used to obtain representations from single and multiple data sources and combine information from EMR data and KGs. Furthermore, DCMF can be used to learn representations in both supervised and unsupervised settings. Thus, DCMF offers an effective way of integrating heterogeneous data sources and infusing auxiliary knowledge into patient representations.
25
Gong, Letian, Youfang Lin, Shengnan Guo, Yan Lin, Tianyi Wang, Erwen Zheng, Zeyu Zhou, and Huaiyu Wan. "Contrastive Pre-training with Adversarial Perturbations for Check-In Sequence Representation Learning." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 4 (June 26, 2023): 4276–83. http://dx.doi.org/10.1609/aaai.v37i4.25546.
Повний текст джерелаАнотація:
A core step of mining human mobility data is to learn accurate representations for user-generated check-in sequences. The learned representations should be able to fully describe the spatial-temporal mobility patterns of users and the high-level semantics of traveling. However, existing check-in sequence representation learning is usually implicitly achieved by end-to-end models designed for specific downstream tasks, resulting in unsatisfactory generalizable abilities and poor performance. Besides, although the sequence representation learning models that follow the contrastive learning pre-training paradigm have achieved breakthroughs in many fields like NLP, they fail to simultaneously consider the unique spatial-temporal characteristics of check-in sequences and need manual adjustments on the data augmentation strategies. So, directly applying them to check-in sequences cannot yield a meaningful pretext task. To this end, in this paper we propose a contrastive pre-training model with adversarial perturbations for check-in sequence representation learning (CACSR). Firstly, we design a novel spatial-temporal augmentation block for disturbing the spatial-temporal features of check-in sequences in the latent space to relieve the stress of designing manual data augmentation strategies. Secondly, to construct an effective contrastive pretext task, we generate “hard” positive and negative pairs for the check-in sequence by adversarial training. These two designs encourage the model to capture the high-level spatial-temporal patterns and semantics of check-in sequences while ignoring the noisy and unimportant details. We demonstrate the effectiveness and versatility of CACSR on two kinds of downstream tasks using three real-world datasets. The results show that our model outperforms both the state-of-the-art pre-training methods and the end-to-end models.
26
Lu, Yuxun, and Ryutaro Ichise. "ProtoE: Enhancing Knowledge Graph Completion Models with Unsupervised Type Representation Learning." Information 13, no. 8 (July 25, 2022): 354. http://dx.doi.org/10.3390/info13080354.
Повний текст джерелаАнотація:
Knowledge graph completion (KGC) models are a feasible approach for manipulating facts in knowledge graphs. However, the lack of entity types in current KGC models results in inaccurate link prediction results. Most existing type-aware KGC models require entity type annotations, which are not always available and expensive to obtain. We propose ProtoE, an unsupervised method for learning implicit type and type constraint representations. ProtoE enhances type-agnostic KGC models by relation-specific prototype embeddings. Our method does not rely on entity type annotations to capture the type and type constraints of entities. Unlike existing unsupervised type representation learning methods, which have only a single representation for entity-type and relation-type constraints, our method can capture multiple type constraints in relations. Experimental results show that our method can improve the performance of both bilinear and translational KGC models in the link prediction task.
27
Qin, Libo, Fuxuan Wei, Minheng Ni, Yue Zhang, Wanxiang Che, Yangming Li, and Ting Liu. "Multi-domain Spoken Language Understanding Using Domain- and Task-aware Parameterization." ACM Transactions on Asian and Low-Resource Language Information Processing 21, no. 4 (July 31, 2022): 1–17. http://dx.doi.org/10.1145/3502198.
Повний текст джерелаАнотація:
Spoken language understanding (SLU) has been addressed as a supervised learning problem, where a set of training data is available for each domain. However, annotating data for a new domain can be both financially costly and non-scalable. One existing approach solves the problem by conducting multi-domain learning where parameters are shared for joint training across domains, which is domain-agnostic and task-agnostic . In the article, we propose to improve the parameterization of this method by using domain-specific and task-specific model parameters for fine-grained knowledge representation and transfer. Experiments on five domains show that our model is more effective for multi-domain SLU and obtain the best results. In addition, we show its transferability when adapting to a new domain with little data, outperforming the prior best model by 12.4%. Finally, we explore the strong pre-trained model in our framework and find that the contributions from our framework do not fully overlap with contextualized word representations (RoBERTa).
28
Reinert, Sandra, Mark Hübener, Tobias Bonhoeffer, and Pieter M. Goltstein. "Mouse prefrontal cortex represents learned rules for categorization." Nature 593, no. 7859 (April 21, 2021): 411–17. http://dx.doi.org/10.1038/s41586-021-03452-z.
Повний текст джерелаАнотація:
AbstractThe ability to categorize sensory stimuli is crucial for an animal’s survival in a complex environment. Memorizing categories instead of individual exemplars enables greater behavioural flexibility and is computationally advantageous. Neurons that show category selectivity have been found in several areas of the mammalian neocortex1–4, but the prefrontal cortex seems to have a prominent role4,5 in this context. Specifically, in primates that are extensively trained on a categorization task, neurons in the prefrontal cortex rapidly and flexibly represent learned categories6,7. However, how these representations first emerge in naive animals remains unexplored, leaving it unclear whether flexible representations are gradually built up as part of semantic memory or assigned more or less instantly during task execution8,9. Here we investigate the formation of a neuronal category representation throughout the entire learning process by repeatedly imaging individual cells in the mouse medial prefrontal cortex. We show that mice readily learn rule-based categorization and generalize to novel stimuli. Over the course of learning, neurons in the prefrontal cortex display distinct dynamics in acquiring category selectivity and are differentially engaged during a later switch in rules. A subset of neurons selectively and uniquely respond to categories and reflect generalization behaviour. Thus, a category representation in the mouse prefrontal cortex is gradually acquired during learning rather than recruited ad hoc. This gradual process suggests that neurons in the medial prefrontal cortex are part of a specific semantic memory for visual categories.
29
Jung, Jinhong, Jaemin Yoo, and U. Kang. "Signed random walk diffusion for effective representation learning in signed graphs." PLOS ONE 17, no. 3 (March 17, 2022): e0265001. http://dx.doi.org/10.1371/journal.pone.0265001.
Повний текст джерелаАнотація:
How can we model node representations to accurately infer the signs of missing edges in a signed social graph? Signed social graphs have attracted considerable attention to model trust relationships between people. Various representation learning methods such as network embedding and graph convolutional network (GCN) have been proposed to analyze signed graphs. However, existing network embedding models are not end-to-end for a specific task, and GCN-based models exhibit a performance degradation issue when their depth increases. In this paper, we propose Signed Diffusion Network (SidNet), a novel graph neural network that achieves end-to-end node representation learning for link sign prediction in signed social graphs. We propose a new random walk based feature aggregation, which is specially designed for signed graphs, so that SidNet effectively diffuses hidden node features and uses more information from neighboring nodes. Through extensive experiments, we show that SidNet significantly outperforms state-of-the-art models in terms of link sign prediction accuracy.
30
Chambers, Claire, Hugo Fernandes, and Konrad Paul Kording. "Policies or knowledge: priors differ between a perceptual and sensorimotor task." Journal of Neurophysiology 121, no. 6 (June 1, 2019): 2267–75. http://dx.doi.org/10.1152/jn.00035.2018.
Повний текст джерелаАнотація:
If the brain abstractly represents probability distributions as knowledge, then the modality of a decision, e.g., movement vs. perception, should not matter. If, on the other hand, learned representations are policies, they may be specific to the task where learning takes place. Here, we test this by asking whether a learned spatial prior generalizes from a sensorimotor estimation task to a two-alternative-forced choice (2-Afc) perceptual comparison task. A model and simulation-based analysis revealed that while participants learn prior distribution in the sensorimotor estimation task, measured priors are consistently broader than sensorimotor priors in the 2-Afc task. That the prior does not fully generalize suggests that sensorimotor priors are more like policies than knowledge. In disagreement with standard Bayesian thought, the modality of the decision has a strong influence on the implied prior distributions. NEW & NOTEWORTHY We do not know whether the brain represents abstract and generalizable knowledge or task-specific policies that map internal states to actions. We find that learning in a sensorimotor task does not generalize strongly to a perceptual task, suggesting that humans learned policies and did not truly acquire knowledge. Priors differ across tasks, thus casting doubt on the central tenet of many Bayesian models, that the brain’s representation of the world is built on generalizable knowledge.
31
Russo, Alessio, and Alexandre Proutiere. "On the Sample Complexity of Representation Learning in Multi-Task Bandits with Global and Local Structure." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 8 (June 26, 2023): 9658–67. http://dx.doi.org/10.1609/aaai.v37i8.26155.
Повний текст джерелаАнотація:
We investigate the sample complexity of learning the optimal arm for multi-task bandit problems. Arms consist of two components: one that is shared across tasks (that we call representation) and one that is task-specific (that we call predictor). The objective is to learn the optimal (representation, predictor)-pair for each task, under the assumption that the optimal representation is common to all tasks. Within this framework, efficient learning algorithms should transfer knowledge across tasks. We consider the best-arm identification problem with fixed confidence, where, in each round, the learner actively selects both a task, and an arm, and observes the corresponding reward. We derive instance-specific sample complexity lower bounds, which apply to any algorithm that identifies the best representation, and the best predictor for a task, with prescribed confidence levels. We devise an algorithm, OSRL-SC, that can learn the optimal representation, and the optimal predictors, separately, and whose sample complexity approaches the lower bound. Theoretical and numerical results demonstrate that OSRL-SC achieves a better scaling with respect to the number of tasks compared to the classical best-arm identification algorithm. The code can be found here https://github.com/rssalessio/OSRL-SC.
32
Liu, Fenglin, Xian Wu, Shen Ge, Wei Fan, and Yuexian Zou. "Federated Learning for Vision-and-Language Grounding Problems." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 07 (April 3, 2020): 11572–79. http://dx.doi.org/10.1609/aaai.v34i07.6824.
Повний текст джерелаАнотація:
Recently, vision-and-language grounding problems, e.g., image captioning and visual question answering (VQA), has attracted extensive interests from both academic and industrial worlds. However, given the similarity of these tasks, the efforts to obtain better results by combining the merits of their algorithms are not well studied. Inspired by the recent success of federated learning, we propose a federated learning framework to obtain various types of image representations from different tasks, which are then fused together to form fine-grained image representations. The representations merge useful features from different vision-and-language grounding problems, and are thus much more powerful than the original representations alone in individual tasks. To learn such image representations, we propose the Aligning, Integrating and Mapping Network (aimNet). The aimNet is validated on three federated learning settings, which include horizontal federated learning, vertical federated learning, and federated transfer learning. Experiments of aimNet-based federated learning framework on two representative tasks, i.e., image captioning and VQA, demonstrate the effective and universal improvements of all metrics over the baselines. In image captioning, we are able to get 14% and 13% relative gain on the task-specific metrics CIDEr and SPICE, respectively. In VQA, we could also boost the performance of strong baselines by up to 3%.
33
Anderson, Brian A., and Haena Kim. "On the representational nature of value-driven spatial attentional biases." Journal of Neurophysiology 120, no. 5 (November 1, 2018): 2654–58. http://dx.doi.org/10.1152/jn.00489.2018.
Повний текст джерелаАнотація:
Reward learning biases attention toward both reward-associated objects and reward-associated regions of space. The relationship between objects and space in the value-based control of attention, as well as the contextual specificity of space-reward pairings, remains unclear. In the present study, using a free-viewing task, we provide evidence of overt attentional biases toward previously rewarded regions of texture scenes that lack objects. When scrutinizing a texture scene, participants look more frequently toward, and spend a longer amount of time looking at, regions that they have repeatedly oriented to in the past as a result of performance feedback. These biases were scene specific, such that different spatial contexts produced different patterns of habitual spatial orienting. Our findings indicate that reinforcement learning can modify looking behavior via a representation that is purely spatial in nature in a context-specific manner. NEW & NOTEWORTHY The representational nature of space in the value-driven control of attention remains unclear. Here, we provide evidence for scene-specific overt spatial attentional biases following reinforcement learning, even though the scenes contained no objects. Our findings indicate that reinforcement learning can modify looking behavior via a representation that is purely spatial in nature in a context-specific manner.
34
Wang, Zhongming, Jiahui Dong, Lianlian Wu, Chong Dai, Jing Wang, Yuqi Wen, Yixin Zhang, Xiaoxi Yang, Song He, and Xiaochen Bo. "DEML: Drug Synergy and Interaction Prediction Using Ensemble-Based Multi-Task Learning." Molecules 28, no. 2 (January 14, 2023): 844. http://dx.doi.org/10.3390/molecules28020844.
Повний текст джерелаАнотація:
Synergistic drug combinations have demonstrated effective therapeutic effects in cancer treatment. Deep learning methods accelerate identification of novel drug combinations by reducing the search space. However, potential adverse drug–drug interactions (DDIs), which may increase the risks for combination therapy, cannot be detected by existing computational synergy prediction methods. We propose DEML, an ensemble-based multi-task neural network, for the simultaneous optimization of five synergy regression prediction tasks, synergy classification, and DDI classification tasks. DEML uses chemical and transcriptomics information as inputs. DEML adapts the novel hybrid ensemble layer structure to construct higher order representation using different perspectives. The task-specific fusion layer of DEML joins representations for each task using a gating mechanism. For the Loewe synergy prediction task, DEML overperforms the state-of-the-art synergy prediction method with an improvement of 7.8% and 13.2% for the root mean squared error and the R2 correlation coefficient. Owing to soft parameter sharing and ensemble learning, DEML alleviates the multi-task learning ‘seesaw effect’ problem and shows no performance loss on other tasks. DEML has a superior ability to predict drug pairs with high confidence and less adverse DDIs. DEML provides a promising way to guideline novel combination therapy strategies for cancer treatment.
35
Song, Kaisong, Yangyang Kang, Jiawei Liu, Xurui Li, Changlong Sun, and Xiaozhong Liu. "A Speaker Turn-Aware Multi-Task Adversarial Network for Joint User Satisfaction Estimation and Sentiment Analysis." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 11 (June 26, 2023): 13582–90. http://dx.doi.org/10.1609/aaai.v37i11.26592.
Повний текст джерелаАнотація:
User Satisfaction Estimation is an important task and increasingly being applied in goal-oriented dialogue systems to estimate whether the user is satisfied with the service. It is observed that whether the user’s needs are met often triggers various sentiments, which can be pertinent to the successful estimation of user satisfaction, and vice versa. Thus, User Satisfaction Estimation (USE) and Sentiment Analysis (SA) should be treated as a joint, collaborative effort, considering the strong connections between the sentiment states of speakers and the user satisfaction. Existing joint learning frameworks mainly unify the two highly pertinent tasks over cascade or shared-bottom implementations, however they fail to distinguish task-specific and common features, which will produce sub-optimal utterance representations for downstream tasks. In this paper, we propose a novel Speaker Turn-Aware Multi-Task Adversarial Network (STMAN) for dialogue-level USE and utterance-level SA. Specifically, we first introduce a multi-task adversarial strategy which trains a task discriminator to make utterance representation more task-specific, and then utilize a speaker-turn aware multi-task interaction strategy to extract the common features which are complementary to each task. Extensive experiments conducted on two real-world service dialogue datasets show that our model outperforms several state-of-the-art methods.
36
León, Fabian, and Fabio Martínez. "A multitask deep representation for Gleason score classification to support grade annotations." Biomedical Physics & Engineering Express 8, no. 3 (April 8, 2022): 035021. http://dx.doi.org/10.1088/2057-1976/ac60c4.
Повний текст джерелаАнотація:
Abstract The Gleason grade system is the main standard to quantify the aggressiveness and progression of prostate cancer. Currently, exists a high disagreement among experts in the diagnosis and stratification of this disease. Deep learning models have emerged as an alternative to classify and support experts automatically. However, these models are limited to learn a rigid stratification rule that can be biased during training to a specific observer. Therefore, this work introduces an embedding representation that integrates an auxiliary task learning to deal with the high inter and intra appearance of the Gleason system. The proposed strategy implements as a main task a triplet loss scheme that builds a feature embedding space with respect to batches of positive and negative histological training patches. As an auxiliary task is added a cross-entropy that helps with inter-class variability of samples while adding robust representations to the main task. The proposed approach shows promising results achieving an average accuracy of 66% and 64%, for two experts without statistical difference. Additionally, reach and average accuracy of 73% in patches where both pathologists are agree, showing the robustness patterns learning from the approach.
37
Xu, Yao, Xueshuang Xiang, and Meiyu Huang. "Task-Driven Common Representation Learning via Bridge Neural Network." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 5573–80. http://dx.doi.org/10.1609/aaai.v33i01.33015573.
Повний текст джерелаАнотація:
This paper introduces a novel deep learning based method, named bridge neural network (BNN) to dig the potential relationship between two given data sources task by task. The proposed approach employs two convolutional neural networks that project the two data sources into a feature space to learn the desired common representation required by the specific task. The training objective with artificial negative samples is introduced with the ability of mini-batch training and it’s asymptotically equivalent to maximizing the total correlation of the two data sources, which is verified by the theoretical analysis. The experiments on the tasks, including pair matching, canonical correlation analysis, transfer learning, and reconstruction demonstrate the state-of-the-art performance of BNN, which may provide new insights into the aspect of common representation learning.
38
Trofimov, Assya, Joseph Paul Cohen, Yoshua Bengio, Claude Perreault, and Sébastien Lemieux. "Factorized embeddings learns rich and biologically meaningful embedding spaces using factorized tensor decomposition." Bioinformatics 36, Supplement_1 (July 1, 2020): i417—i426. http://dx.doi.org/10.1093/bioinformatics/btaa488.
Повний текст джерелаАнотація:
Abstract Motivation The recent development of sequencing technologies revolutionized our understanding of the inner workings of the cell as well as the way disease is treated. A single RNA sequencing (RNA-Seq) experiment, however, measures tens of thousands of parameters simultaneously. While the results are information rich, data analysis provides a challenge. Dimensionality reduction methods help with this task by extracting patterns from the data by compressing it into compact vector representations. Results We present the factorized embeddings (FE) model, a self-supervised deep learning algorithm that learns simultaneously, by tensor factorization, gene and sample representation spaces. We ran the model on RNA-Seq data from two large-scale cohorts and observed that the sample representation captures information on single gene and global gene expression patterns. Moreover, we found that the gene representation space was organized such that tissue-specific genes, highly correlated genes as well as genes participating in the same GO terms were grouped. Finally, we compared the vector representation of samples learned by the FE model to other similar models on 49 regression tasks. We report that the representations trained with FE rank first or second in all of the tasks, surpassing, sometimes by a considerable margin, other representations. Availability and implementation A toy example in the form of a Jupyter Notebook as well as the code and trained embeddings for this project can be found at: https://github.com/TrofimovAssya/FactorizedEmbeddings. Supplementary information Supplementary data are available at Bioinformatics online.
39
Zhang, Liyi, Zengguang Tian, Yi Tang, and Zuo Jiang. "Task-Covariant Representations for Few-Shot Learning on Remote Sensing Images." Mathematics 11, no. 8 (April 19, 2023): 1930. http://dx.doi.org/10.3390/math11081930.
Повний текст джерелаАнотація:
In the regression and classification of remotely sensed images through meta-learning, techniques exploit task-invariant information to quickly adapt to new tasks with fewer gradient updates. Despite its usefulness, task-invariant information alone may not effectively capture task-specific knowledge, leading to reduced model performance on new tasks. As a result, the concept of task-covariance has gained significant attention from researchers. We propose task-covariant representations for few-shot Learning on remote sensing images that utilizes capsule networks to effectively represent the covariance relationships among objects. This approach is motivated by the superior ability of capsule networks to capture such relationships. To capture and leverage the covariance relations between tasks, we employ vector capsules and adapt our model parameters based on the newly learned task covariance relations. Our proposed meta-learning algorithm offers a novel approach to effectively address the real task distribution by incorporating both general and specific task information. Based on the experimental results, our proposed meta-learning algorithm shows a significant improvement in both the average accuracy and training efficiency compared to the best model in the experiments. On average, the algorithm increases the accuracy by approximately 4% and improves the training efficiency by approximately 8%.
40
DUNN, JONATHAN. "Computational learning of construction grammars." Language and Cognition 9, no. 2 (March 28, 2016): 254–92. http://dx.doi.org/10.1017/langcog.2016.7.
Повний текст джерелаАнотація:
abstractThis paper presents an algorithm for learning the construction grammar of a language from a large corpus. This grammar induction algorithm has two goals: first, to show that construction grammars are learnable without highly specified innate structure; second, to develop a model of which units do or do not constitute constructions in a given dataset. The basic task of construction grammar induction is to identify the minimum set of constructions that represents the language in question with maximum descriptive adequacy. These constructions must (1) generalize across an unspecified number of units while (2) containing mixed levels of representation internally (e.g., both item-specific and schematized representations), and (3) allowing for unfilled and partially filled slots. Additionally, these constructions may (4) contain recursive structure within a given slot that needs to be reduced in order to produce a sufficiently schematic representation. In other words, these constructions are multi-length, multi-level, possibly discontinuous co-occurrences which generalize across internal recursive structures. These co-occurrences are modeled using frequency and the ΔP measure of association, expanded in novel ways to cover multi-unit sequences. This work provides important new evidence for the learnability of construction grammars as well as a tool for the automated corpus analysis of constructions.
41
Guo, Yiyou, and Chao Wei. "Multi-Task Learning Using Gradient Balance and Clipping with an Application in Joint Disparity Estimation and Semantic Segmentation." Electronics 11, no. 8 (April 12, 2022): 1217. http://dx.doi.org/10.3390/electronics11081217.
Повний текст джерелаАнотація:
In this paper, we propose a novel multi-task learning (MTL) strategy from the gradient optimization view which enables automatically learning the optimal gradient from different tasks. In contrast with current multi-task learning methods which rely on careful network architecture adjustment or elaborate loss functions optimization, the proposed gradient-based MTL is simple and flexible. Specifically, we introduce a multi-task stochastic gradient descent optimization (MTSGD) to learn task-specific and shared representation in the deep neural network. In MTSGD, we decompose the total gradient into multiple task-specific sub-gradients and find the optimal sub-gradient via gradient balance and clipping operations. In this way, the learned network can satisfy the performance of specific task optimization while maintaining the shared representation. We take the joint learning of semantic segmentation and disparity estimation tasks as the exemplar to verify the effectiveness of the proposed method. Extensive experimental results on a large-scale dataset show that our proposed algorithm is superior to the baseline methods by a large margin. Meanwhile, we perform a series of ablation studies to have a deep analysis of gradient descent for MTL.
42
Lampinen, Andrew K., and James L. McClelland. "Transforming task representations to perform novel tasks." Proceedings of the National Academy of Sciences 117, no. 52 (December 10, 2020): 32970–81. http://dx.doi.org/10.1073/pnas.2008852117.
Повний текст джерелаАнотація:
An important aspect of intelligence is the ability to adapt to a novel task without any direct experience (zero shot), based on its relationship to previous tasks. Humans can exhibit this cognitive flexibility. By contrast, models that achieve superhuman performance in specific tasks often fail to adapt to even slight task alterations. To address this, we propose a general computational framework for adapting to novel tasks based on their relationship to prior tasks. We begin by learning vector representations of tasks. To adapt to new tasks, we propose metamappings, higher-order tasks that transform basic task representations. We demonstrate the effectiveness of this framework across a wide variety of tasks and computational paradigms, ranging from regression to image classification and reinforcement learning. We compare to both human adaptability and language-based approaches to zero-shot learning. Across these domains, metamapping is successful, often achieving 80 to 90% performance, without any data, on a novel task, even when the new task directly contradicts prior experience. We further show that metamapping can not only generalize to new tasks via learned relationships, but can also generalize using novel relationships unseen during training. Finally, using metamapping as a starting point can dramatically accelerate later learning on a new task and reduce learning time and cumulative error substantially. Our results provide insight into a possible computational basis of intelligent adaptability and offer a possible framework for modeling cognitive flexibility and building more flexible artificial intelligence systems.
43
Sitaula, Chiranjibi, Anish Basnet, and Sunil Aryal. "Vector representation based on a supervised codebook for Nepali documents classification." PeerJ Computer Science 7 (March 3, 2021): e412. http://dx.doi.org/10.7717/peerj-cs.412.
Повний текст джерелаАнотація:
Document representation with outlier tokens exacerbates the classification performance due to the uncertain orientation of such tokens. Most existing document representation methods in different languages including Nepali mostly ignore the strategies to filter them out from documents before learning their representations. In this article, we propose a novel document representation method based on a supervised codebook to represent the Nepali documents, where our codebook contains only semantic tokens without outliers. Our codebook is domain-specific as it is based on tokens in a given corpus that have higher similarities with the class labels in the corpus. Our method adopts a simple yet prominent representation method for each word, called probability-based word embedding. To show the efficacy of our method, we evaluate its performance in the document classification task using Support Vector Machine and validate against widely used document representation methods such as Bag of Words, Latent Dirichlet allocation, Long Short-Term Memory, Word2Vec, Bidirectional Encoder Representations from Transformers and so on, using four Nepali text datasets (we denote them shortly as A1, A2, A3 and A4). The experimental results show that our method produces state-of-the-art classification performance (77.46% accuracy on A1, 67.53% accuracy on A2, 80.54% accuracy on A3 and 89.58% accuracy on A4) compared to the widely used existing document representation methods. It yields the best classification accuracy on three datasets (A1, A2 and A3) and a comparable accuracy on the fourth dataset (A4). Furthermore, we introduce the largest Nepali document dataset (A4), called NepaliLinguistic dataset, to the linguistic community.
44
Jaiswal, Mimansa, and Emily Mower Provost. "Privacy Enhanced Multimodal Neural Representations for Emotion Recognition." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 05 (April 3, 2020): 7985–93. http://dx.doi.org/10.1609/aaai.v34i05.6307.
Повний текст джерелаАнотація:
Many mobile applications and virtual conversational agents now aim to recognize and adapt to emotions. To enable this, data are transmitted from users' devices and stored on central servers. Yet, these data contain sensitive information that could be used by mobile applications without user's consent or, maliciously, by an eavesdropping adversary. In this work, we show how multimodal representations trained for a primary task, here emotion recognition, can unintentionally leak demographic information, which could override a selected opt-out option by the user. We analyze how this leakage differs in representations obtained from textual, acoustic, and multimodal data. We use an adversarial learning paradigm to unlearn the private information present in a representation and investigate the effect of varying the strength of the adversarial component on the primary task and on the privacy metric, defined here as the inability of an attacker to predict specific demographic information. We evaluate this paradigm on multiple datasets and show that we can improve the privacy metric while not significantly impacting the performance on the primary task. To the best of our knowledge, this is the first work to analyze how the privacy metric differs across modalities and how multiple privacy concerns can be tackled while still maintaining performance on emotion recognition.
45
Cerda, Vanessa R., Paola Montufar Soria, and Nicole Y. Wicha. "Reevaluating the Language of Learning Advantage in Bilingual Arithmetic: An ERP Study on Spoken Multiplication Verification." Brain Sciences 12, no. 5 (April 21, 2022): 532. http://dx.doi.org/10.3390/brainsci12050532.
Повний текст джерелаАнотація:
Many studies of bilingual arithmetic report better performance when verifying arithmetic facts in the language of learning (LA+) over the other language (LA−). This could be due to language-specific memory representations, processes established during learning, or to language and task factors not related to math. The current study builds on a small number of event-related potential (ERP) studies to test this question while controlling language proficiency and eliminating potential task confounds. Adults proficient in two languages verified single-digit multiplications presented as spoken number words in LA+ and LA−, separately. ERPs and correctness judgments were measured from solution onset. Equivalent P300 effects, with larger positive amplitude for correct than incorrect solutions, were observed in both languages (Experiment 1A), even when stimuli presentation rate was shortened to increase difficulty (Experiment 1B). This effect paralleled the arithmetic correctness effect for trials presented as all digits (e.g., 2 4 8 versus 2 4 10), reflecting efficient categorization of the solutions, and was distinct from an N400 generated in a word–picture matching task, reflecting meaning processing (Experiment 2). The findings reveal that the language effects on arithmetic are likely driven by language and task factors rather than differences in memory representation in each language.
46
Lu, Wenpeng, Rui Yu, Shoujin Wang, Can Wang, Ping Jian, and Heyan Huang. "Sentence Semantic Matching Based on 3D CNN for Human–Robot Language Interaction." ACM Transactions on Internet Technology 21, no. 4 (July 16, 2021): 1–24. http://dx.doi.org/10.1145/3450520.
Повний текст джерелаАнотація:
The development of cognitive robotics brings an attractive scenario where humans and robots cooperate to accomplish specific tasks. To facilitate this scenario, cognitive robots are expected to have the ability to interact with humans with natural language, which depends on natural language understanding ( NLU ) technologies. As one core task in NLU, sentence semantic matching ( SSM ) has widely existed in various interaction scenarios. Recently, deep learning–based methods for SSM have become predominant due to their outstanding performance. However, each sentence consists of a sequence of words, and it is usually viewed as one-dimensional ( 1D ) text, leading to the existing available neural models being restricted into 1D sequential networks. A few researches attempt to explore the potential of 2D or 3D neural models in text representation. However, it is hard for their works to capture the complex features in texts, and thus the achieved performance improvement is quite limited. To tackle this challenge, we devise a novel 3D CNN-based SSM ( 3DSSM ) method for human–robot language interaction. Specifically, first, a specific architecture called feature cube network is designed to transform a 1D sentence into a multi-dimensional representation named as semantic feature cube. Then, a 3D CNN module is employed to learn a semantic representation for the semantic feature cube by capturing both the local features embedded in word representations and the sequential information among successive words in a sentence. Given a pair of sentences, their representations are concatenated together to feed into another 3D CNN to capture the interactive features between them to generate the final matching representation. Finally, the semantic matching degree is judged with the sigmoid function by taking the learned matching representation as the input. Extensive experiments on two real-world datasets demonstrate that 3DSSM is able to achieve comparable or even better performance over the state-of-the-art competing methods.
47
Zeng, Te, and Francis C. M. Lau. "Automatic Melody Harmonization via Reinforcement Learning by Exploring Structured Representations for Melody Sequences." Electronics 10, no. 20 (October 11, 2021): 2469. http://dx.doi.org/10.3390/electronics10202469.
Повний текст джерелаАнотація:
We present a novel reinforcement learning architecture that learns a structured representation for use in symbolic melody harmonization. Probabilistic models are predominant in melody harmonization tasks, most of which only treat melody notes as independent observations and do not take note of substructures in the melodic sequence. To fill this gap, we add substructure discovery as a crucial step in automatic chord generation. The proposed method consists of a structured representation module that generates hierarchical structures for the symbolic melodies, a policy module that learns to break a melody into segments (whose boundaries concur with chord changes) and phrases (the subunits in segments) and a harmonization module that generates chord sequences for each segment. We formulate the structure discovery process as a sequential decision problem with a policy gradient RL method selecting the boundary of each segment or phrase to obtain an optimized structure. We conduct experiments on our preprocessed HookTheory Lead Sheet Dataset, which has 17,979 melody/chord pairs. The results demonstrate that our proposed method can learn task-specific representations and, thus, yield competitive results compared with state-of-the-art baselines.
48
Clegg, Benjamin A. "Stimulus-Specific Sequence Representation in Serial Reaction Time Tasks." Quarterly Journal of Experimental Psychology Section A 58, no. 6 (August 2005): 1087–101. http://dx.doi.org/10.1080/02724980443000485.
Повний текст джерелаАнотація:
Some recent evidence has favoured purely response-based implicit representation of sequences in serial reaction time tasks. Three experiments were conducted using serial reaction time tasks featuring four spatial stimuli mapped in categories to two responses. Deviant items from the expected sequence that required the expected response resulted in increased response latencies. The findings demonstrated a stimulus-specific form of representation that operates in the serial reaction time task. No evidence was found to suggest that the stimulus-specific learning was contingent on explicit knowledge of the sequence. Such stimulus-based learning would be congruent with a shortcut within an information-processing framework and, combined with other research findings, suggests that there are multiple loci for learning effects.
49
Vangheluwe, Sophie, Nicole Wenderoth, and Stephan P. Swinnen. "Learning and Transfer of an Ipsilateral Coordination Task: Evidence for a Dual-layer Movement Representation." Journal of Cognitive Neuroscience 17, no. 9 (September 2005): 1460–70. http://dx.doi.org/10.1162/0898929054985392.
Повний текст джерелаАнотація:
The present study addressed the nature of the memory representation for interlimb coordination tasks. For this purpose, the acquisition of a multifrequency (2:1) task with the ipsilateral limbs and transfer to the ipsilateral and contralateral body side was examined. In particular, subjects practiced a 2:1 coordination pattern whereby the right arm moved twice as fast as the right leg, or vice versa. Subsequently, they transferred the practiced 2:1 task to three different conditions: (1) the converse partner (i.e., the slow-moving limb had to move fast, and vice versa) at the ipsilateral body side, and (2) the identical and (3) converse 2:1 pattern at the contralateral body side. Findings revealed positive transfer of the identical and converse 2:1 pattern to the contralateral body side. However, no transfer of the learned pattern to its converse partner at the same body side was revealed. We propose a new memory representation model for coordination patterns, composed of an effector-independent and effector-specific component (dual-layer model). It is hypothesized that the general movement goal (i.e., moving one limb twice as fast as the other) constitutes the abstract, higher-level representation that may account for positive contralateral transfer. Conversely, the effector-specific component contains task-specific lower-level muscle synergies that are acquired through practice, prohibiting positive transfer when shifting task allocation within the same effectors. These findings are consistent with recent neuroscientific evidence for neuroplastic changes in distributed brain areas.
50
Chen, Chen, Yuchen Hu, Qiang Zhang, Heqing Zou, Beier Zhu, and Eng Siong Chng. "Leveraging Modality-Specific Representations for Audio-Visual Speech Recognition via Reinforcement Learning." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 11 (June 26, 2023): 12607–15. http://dx.doi.org/10.1609/aaai.v37i11.26484.
Повний текст джерелаАнотація:
Audio-visual speech recognition (AVSR) has gained remarkable success for ameliorating the noise-robustness of speech recognition. Mainstream methods focus on fusing audio and visual inputs to obtain modality-invariant representations. However, such representations are prone to over-reliance on audio modality as it is much easier to recognize than video modality in clean conditions. As a result, the AVSR model underestimates the importance of visual stream in face of noise corruption. To this end, we leverage visual modality-specific representations to provide stable complementary information for the AVSR task. Specifically, we propose a reinforcement learning (RL) based framework called MSRL, where the agent dynamically harmonizes modality-invariant and modality-specific representations in the auto-regressive decoding process. We customize a reward function directly related to task-specific metrics (i.e., word error rate), which encourages the MSRL to effectively explore the optimal integration strategy. Experimental results on the LRS3 dataset show that the proposed method achieves state-of-the-art in both clean and various noisy conditions. Furthermore, we demonstrate the better generality of MSRL system than other baselines when test set contains unseen noises.