Relevant bibliographies by topics / Spurious features

Contents

  1. Journal articles
  2. Books
  3. Book chapters
  4. Conference papers

Academic literature on the topic 'Spurious features'

Author: Grafiati
Published: 25 November 2024

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles

Select a source type:

Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Spurious features.'

Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.

You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.

Journal articles on the topic "Spurious features"

1

Chen, Kaitao, Shiliang Sun, and Jing Zhao. "CaMIL: Causal Multiple Instance Learning for Whole Slide Image Classification." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 2 (March 24, 2024): 1120–28. http://dx.doi.org/10.1609/aaai.v38i2.27873.

Full text
Abstract:
Whole slide image (WSI) classification is a crucial component in automated pathology analysis. Due to the inherent challenges of high-resolution WSIs and the absence of patch-level labels, most of the proposed methods follow the multiple instance learning (MIL) formulation. While MIL has been equipped with excellent instance feature extractors and aggregators, it is prone to learn spurious associations that undermine the performance of the model. For example, relying solely on color features may lead to erroneous diagnoses due to spurious associations between the disease and the color of patches. To address this issue, we develop a causal MIL framework for WSI classification, effectively distinguishing between causal and spurious associations. Specifically, we use the expectation of the intervention P(Y | do(X)) for bag prediction rather than the traditional likelihood P(Y | X). By applying the front-door adjustment, the spurious association is effectively blocked, where the intervened mediator is aggregated from patch-level features. We evaluate our proposed method on two publicly available WSI datasets, Camelyon16 and TCGA-NSCLC. Our causal MIL framework shows outstanding performance and is plug-and-play, seamlessly integrating with various feature extractors and aggregators.
APA, Harvard, Vancouver, ISO, and other styles
2

Su, Donglin, Qian Shi, Hui Xu, and Wang Wang. "Nonintrusive Load Monitoring Based on Complementary Features of Spurious Emissions." Electronics 8, no. 9 (September 7, 2019): 1002. http://dx.doi.org/10.3390/electronics8091002.

Full text
Abstract:
In this paper, a novel method that utilizes the fractional correlation-based algorithm and the B-spline curve fitting-based algorithm is proposed to extract the complementary features for detecting the operating states of appliances. The identification of appliance operating states is one of the key parts for nonintrusive load monitoring (NILM). Considering the individual spurious emissions generated because of nonlinear components in each electronic device, the spurious emissions from the power cord can be picked up to solve the problem of data storage. Five types of common household appliances are considered in this study. The fractional correlation-based algorithm and B-spline curve fitting-based algorithm are used to extract two groups of complementary features from the spurious emissions of those five types of appliances. The experimental results show that the feature vectors extracted using the proposed method are obviously distinguishable. In addition, the features extracted show a good long-time stability, which is verified through a five-day experiment. Finally, based on support vector machine (SVM) and Dempster–Shafer (D-S) evidence theory, the identification accuracy reaches 85.5% using a combining classifier incorporated with the features extracted from the proposed methods.
APA, Harvard, Vancouver, ISO, and other styles
3

KARIMI, SAEED, and HAMDİ DİBEKLİOĞLU. "Uncovering and mitigating spurious features in domain generalization." Turkish Journal of Electrical Engineering and Computer Sciences 32, no. 2 (March 14, 2024): 320–37. http://dx.doi.org/10.55730/1300-0632.4071.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Du, Mengnan, Ruixiang Tang, Weijie Fu, and Xia Hu. "Towards Debiasing DNN Models from Spurious Feature Influence." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 9 (June 28, 2022): 9521–28. http://dx.doi.org/10.1609/aaai.v36i9.21185.

Full text
Abstract:
Recent studies indicate that deep neural networks (DNNs) are prone to show discrimination towards certain demographic groups. We observe that algorithmic discrimination can be explained by the high reliance of the models on fairness sensitive features. Motivated by this observation, we propose to achieve fairness by suppressing the DNN models from capturing the spurious correlation between those fairness sensitive features with the underlying task. Specifically, we firstly train a bias-only teacher model which is explicitly encouraged to maximally employ fairness sensitive features for prediction. The teacher model then counter-teaches a debiased student model so that the interpretation of the student model is orthogonal to the interpretation of the teacher model. The key idea is that since the teacher model relies explicitly on fairness sensitive features for prediction, the orthogonal interpretation loss enforces the student network to reduce its reliance on sensitive features and instead capture more task relevant features for prediction. Experimental analysis indicates that our framework substantially reduces the model's attention on fairness sensitive features. Experimental results on four datasets further validate that our framework has consistently improved the fairness with respect to three group fairness metrics, with a comparable or even better accuracy.
APA, Harvard, Vancouver, ISO, and other styles
5

Ming, Yifei, Hang Yin, and Yixuan Li. "On the Impact of Spurious Correlation for Out-of-Distribution Detection." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 9 (June 28, 2022): 10051–59. http://dx.doi.org/10.1609/aaai.v36i9.21244.

Full text
Abstract:
Modern neural networks can assign high confidence to inputs drawn from outside the training distribution, posing threats to models in real-world deployments. While much research attention has been placed on designing new out-of-distribution (OOD) detection methods, the precise definition of OOD is often left in vagueness and falls short of the desired notion of OOD in reality. In this paper, we present a new formalization and model the data shifts by taking into account both the invariant and environmental (spurious) features. Under such formalization, we systematically investigate how spurious correlation in the training set impacts OOD detection. Our results suggest that the detection performance is severely worsened when the correlation between spurious features and labels is increased in the training set. We further show insights on detection methods that are more effective in reducing the impact of spurious correlation, and provide theoretical analysis on why reliance on environmental features leads to high OOD detection error. Our work aims to facilitate better understanding of OOD samples and their formalization, as well as the exploration of methods that enhance OOD detection. Code is available at https://github.com/deeplearning-wisc/Spurious_OOD.
APA, Harvard, Vancouver, ISO, and other styles
6

Popovic, Brankica, and Ljiljana Maskovic. "Fingerprint minutiae filtering based on multiscale directional information." Facta universitatis - series: Electronics and Energetics 20, no. 2 (2007): 233–44. http://dx.doi.org/10.2298/fuee0702233p.

Full text
Abstract:
Automatic identification of humans based on their fingerprints is still one of the most reliable identification methods in criminal and forensic applications, and is widely applied in civil applications as well. Most automatic systems available today use distinctive fingerprint features called minutiae for fingerprint comparison. Conventional feature extraction algorithm can produce a large number of spurious minutiae if fingerprint pattern contains large regions of broken ridges (often called creases). This can drastically reduce the recognition rate in automatic fingerprint identification systems. We can say that for performance of those systems it is more important not to extract spurious (false) minutia even though it means some genuine might be missing as well. In this paper multiscale directional information obtained from orientation field image is used to filter those spurious minutiae, resulting in multiple decrease of their number.
APA, Harvard, Vancouver, ISO, and other styles
7

ARTUSO, Francesco, Francesco FIDECARO, Francesco D'ALESSANDRO, Gino IANNACE, Gaetano LICITRA, Geremia POMPEI, and Luca FREDIANELLI. "Identifying optimal feature sets for acoustic signal classification in environmental noise measurements." INTER-NOISE and NOISE-CON Congress and Conference Proceedings 270, no. 4 (October 4, 2024): 7540–49. http://dx.doi.org/10.3397/in_2024_3974.

Full text
Abstract:
Whatever the sound source to be evaluated, spurious events or unwanted sounds will always be present in environmental noise measurements. Spurious events are not characteristic of standard residual noise and must be removed prior to subsequent analyses. Currently, the removal step is deferred solely to the objective evaluation of the sound pattern and/or spectrogram by an operator. This results in the loss of many man-hours. Machine learning can be used to develop a tool capable of recognizing and removing spurious events in noise measurements. The tool must be able to account for various sounds, whether human-made or animal, and must be applicable to any environmental scenario. This is not a straightforward task, in fact if humans can easily distinguish between two sounds, such as a birds' chirps and a car passing by, based on prior experience, a machine may not be able to do so without apprenticeship. Therefore, a learning methodology must be constructed for the machine by establishing recognizable patterns. The aim of this paper is to identify the feature sets which allow the algorithm to differentiate spurious sounds in the best way. These features will represent the semantic value of the signal.
APA, Harvard, Vancouver, ISO, and other styles
8

Chen, Ziliang, Yongsen Zheng, Zhao-Rong Lai, Quanlong Guan, and Liang Lin. "Diagnosing and Rectifying Fake OOD Invariance: A Restructured Causal Approach." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 10 (March 24, 2024): 11471–79. http://dx.doi.org/10.1609/aaai.v38i10.29028.

Full text
Abstract:
Invariant representation learning (IRL) encourages the prediction from invariant causal features to labels deconfounded from the environments, advancing the technical roadmap of out-of-distribution (OOD) generalization. Despite spotlights around, recent theoretical result verified that some causal features recovered by IRLs merely pretend domain-invariantly in the training environments but fail in unseen domains. The fake invariance severely endangers OOD generalization since the trustful objective can not be diagnosed and existing causal remedies are invalid to rectify. In this paper, we review a IRL family (InvRat) under the Partially and Fully Informative Invariant Feature Structural Causal Models (PIIF SCM /FIIF SCM) respectively, to certify their weaknesses in representing fake invariant features, then, unify their causal diagrams to propose ReStructured SCM (RS-SCM). RS-SCM can ideally rebuild the spurious and the fake invariant features simultaneously. Given this, we further develop an approach based on conditional mutual information with respect to RS-SCM, then rigorously rectify the spurious and fake invariant effects. It can be easily implemented by a small feature selection subnet introduced in the IRL family, which is alternatively optimized to achieve our goal. Experiments verified the superiority of our approach to fight against the fake invariant issue across a variety of OOD generalization benchmarks.
APA, Harvard, Vancouver, ISO, and other styles
9

Wang, Zhao, and Aron Culotta. "Robustness to Spurious Correlations in Text Classification via Automatically Generated Counterfactuals." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 16 (May 18, 2021): 14024–31. http://dx.doi.org/10.1609/aaai.v35i16.17651.

Full text
Abstract:
Spurious correlations threaten the validity of statistical classifiers. While model accuracy may appear high when the test data is from the same distribution as the training data, it can quickly degrade when the test distribution changes. For example, it has been shown that classifiers perform poorly when humans make minor modifications to change the label of an example. One solution to increase model reliability and generalizability is to identify causal associations between features and classes. In this paper, we propose to train a robust text classifier by augmenting the training data with automatically generated counterfactual data. We first identify likely causal features using a statistical matching approach. Next, we generate counterfactual samples for the original training data by substituting causal features with their antonyms and then assigning opposite labels to the counterfactual samples. Finally, we combine the original data and counterfactual data to train a robust classifier. Experiments on two classification tasks show that a traditional classifier trained on the original data does very poorly on human-generated counterfactual samples (e.g., 10%-37% drop in accuracy). However, the classifier trained on the combined data is more robust and performs well on both the original test data and the counterfactual test data (e.g., 12%-25% increase in accuracy compared with the traditional classifier). Detailed analysis shows that the robust classifier makes meaningful and trustworthy predictions by emphasizing causal features and de-emphasizing non-causal features.
APA, Harvard, Vancouver, ISO, and other styles
10

Smy, T., M. Salahuddin, S. K. Dew, and M. J. Brett. "Explanation of spurious features in tungsten deposition using an atomic momentum model." Journal of Applied Physics 78, no. 6 (September 15, 1995): 4157–63. http://dx.doi.org/10.1063/1.359875.

Full text
APA, Harvard, Vancouver, ISO, and other styles
More sources

Books on the topic "Spurious features"

1

Peacocke, Christopher. The Primacy of Metaphysics. Oxford University Press, 2019. http://dx.doi.org/10.1093/oso/9780198835578.001.0001.

Full text
Abstract:
Is the metaphysics of a domain prior in the order of philosophical explanation to a theory of intentional contents and meanings about that domain? Or is the opposite true? This book argues from the nature of meaning and intentional content to the conclusion that content and meaning are never prior to the metaphysics. For every domain, either a metaphysics-first view or a no-priority view is correct. Metaphysics-first views are developed for several specific domains. For extensive magnitudes, a new realistic metaphysics is developed, and this metaphysics is used to explain features of the perception of magnitudes, and to elucidate analogue computation and analogue representation. A metaphysics-first treatment of time is developed and used to develop new accounts of temporal representation, and to address some puzzles about time and present-tense content. A metaphysics-first treatment of subject and the first person develops a new account of the ownership of mental events by subjects, and argues for a greater role of agency in the first person than in earlier accounts. A noncausal metaphysics-first view is developed for the natural numbers and the real numbers. The account gives an explanatory priority to the application of numbers to properties and to ratios of magnitudes. The final chapter of the book argues the materials earlier in the book permit a new account of the limits of intelligibility. Spurious concepts, such as absolute space, are ones for which there is no account of the relation that would have to hold for a thinker to latch onto it.
APA, Harvard, Vancouver, ISO, and other styles

Book chapters on the topic "Spurious features"

1

Chen, Chi-Yu, Pu Ching, Pei-Hsin Huang, and Min-Chun Hu. "Where Are Biases? Adversarial Debiasing with Spurious Feature Visualization." In MultiMedia Modeling, 1–14. Cham: Springer Nature Switzerland, 2024. http://dx.doi.org/10.1007/978-3-031-53305-1_1.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Currie Hall, Daniel. "Contrast and content in phonological features." In Primitives of Phonological Structure, 108–30. Oxford University PressOxford, 2023. http://dx.doi.org/10.1093/oso/9780198791126.003.0005.

Full text
Abstract:
Abstract ‘Substance-free’ theories of phonology take two different approaches to eliminating phonetic information from phonological computations, positing either that distinctive features have phonetic content but phonological rules can manipulate them in arbitrary ways (Hale and Reiss 2003, 2008) or that features are based on phonological patterning and need not have any identifiable phonetic content at all (Blaho 2008; Samuels 2009). This chapter argues that the key insights of substance-free phonology can be maintained in a system that allows a limited role for phonetic substance. Methodologically, requiring that features have phonetic content and rules be formally natural limits the generative power of the system, forcing analysts to look more closely at apparently unnatural rules and classes, and precluding analyses based on spurious generalizations. At the same time, the fact that phonetic content does not always determine phonological patterning can be explained through underspecification based on contrast.
APA, Harvard, Vancouver, ISO, and other styles
3

Iosad, Pavel. "The ATR/Laryngeal connection and emergent features." In Primitives of Phonological Structure, 161–208. Oxford University PressOxford, 2023. http://dx.doi.org/10.1093/oso/9780198791126.003.0007.

Full text
Abstract:
Abstract ‘Substance-free’ theories of phonology take two different approaches to eliminating phonetic information from phonological computations, positing either that distinctive features have phonetic content but phonological rules can manipulate them in arbitrary ways (Hale & Reiss 2003, 2008) or that features are based on phonological patterning and need not have any identifiable phonetic content at all (Blaho 2008; Samuels 2009). This chapter argues that the key insights of substance-free phonology can be maintained in a system that allows a limited role for phonetic substance. Methodologically, requiring that features have phonetic content and rules be formally natural limits the generative power of the system, forcing analysts to look more closely at apparently unnatural rules and classes, and precluding analyses based on spurious generalizations. At the same time, the fact that phonetic content does not always determine phonological patterning can be explained through underspecification based on contrast.
APA, Harvard, Vancouver, ISO, and other styles
4

Hmood, Ali K., and Ching Y. Suen. "An Ensemble of Character Features and Fine-Tuned Convolutional Neural Network for Spurious Coin Detection." In Frontiers in Pattern Recognition and Artificial Intelligence, 169–87. WORLD SCIENTIFIC, 2019. http://dx.doi.org/10.1142/9789811203527_0010.

Full text
APA, Harvard, Vancouver, ISO, and other styles
5

Gibbs, John C. "Moral Development, Moral Identity, and Prosocial Behavior." In Moral Development and Reality, 157–79. Oxford University Press, 2019. http://dx.doi.org/10.1093/oso/9780190878214.003.0006.

Full text
Abstract:
This chapter focuses on some of the variables accounting for individual differences in the likelihood of prosocial behavior. “Prosocial behavior” can range from a particular intervention to a lifetime dedicated to just and good causes. Highly prosocial individuals (moral exemplars) tend to be morally mature and highly empathic but field-independent (Moral Type B, internal locus of control, high self-efficacy) persons who perceive morality as central to their sense of self (high moral identity). Moral identity can join the main primary (affective and cognitive) sources of moral motivation. Finally, to take effective sustained action, even highly prosocial individuals need grit or ego strength, defined in terms of affect-regulating follow-through skills. Distinguishing features of genuine (versus spurious) moral exemplars are considered at the end of the chapter.
APA, Harvard, Vancouver, ISO, and other styles
6

Bag, Soumen. "A Nearest Opposite Contour Pixel Based Thinning Strategy for Character Images." In Advances in Multimedia and Interactive Technologies, 123–40. IGI Global, 2017. http://dx.doi.org/10.4018/978-1-5225-1025-3.ch006.

Full text
Abstract:
Thinning of character images is a big challenge. Removal of strokes or deformities in thinning is a difficult problem. In this paper, we have proposed a nearest opposite contour pixel based thinning strategy used for performing skeletonization of printed and handwritten character images. In this method, we have used shape characteristics of text to get skeleton of nearly same as the true character shape. This approach helps to preserve the local features and true shape of the character images. The proposed algorithm produces one pixel-width thin skeleton. As a by-product of our thinning approach, the skeleton also gets segmented into strokes in vector form. Hence further stroke segmentation is not required. Experiment is done on printed English and Bengali characters and we obtain less spurious branches comparing with other thinning methods without any post-processing.
APA, Harvard, Vancouver, ISO, and other styles
7

Miron, Jeffrey A., and Stephen P. Zeldes. "Seasonality, Cost Shocks, and the Production Smoothing Model of Inventories." In Modelling Seasonality, 209–46. Oxford University PressOxford, 1992. http://dx.doi.org/10.1093/oso/9780198773177.003.0010.

Full text
Abstract:
Abstract Abstract. A great deal of research on the empirical behavior of inventories examines some variant of the production smoothing model of finished goods inventories. The overall assessment of this model that exists in the literature is quite negative: there is little evidence that manufacturers hold inventories of finished goods in order to smooth production patterns. This paper examines whether this negative assessment of the model is due to one or both of two features: cost shocks and seasonal fluctuations. The reason for considering cost shocks is that, if firms are buffeted more by cost shocks than demand shocks, production should optimally be more variable than sales. The reasons for considering seasonal fluctuations are that seasonal fluctuations account for a major portion of the variance in production and sales, that seasonal fluctuations are precisely the kinds of fluctuations that producers should most easily smooth, and that seasonally adjusted data are likely to produce spurious rejections of the production smoothing model even when it is correct.
APA, Harvard, Vancouver, ISO, and other styles
8

Allchin, Douglas. "Genes R Us." In Sacred Bovines. Oxford University Press, 2017. http://dx.doi.org/10.1093/oso/9780190490362.003.0027.

Full text
Abstract:
DNA fingerprints are not prints of fingers. So why the name? The “fingerprint” label, of course, conveys far more than some pattern of swirls, whorls, and arches on the skin. As celebrated in detective lore, fingerprints are emblems of uniqueness. DNA can thus form a “fingerprint” by establishing personal identity. Genes are often characterized as “information.” Thus, DNA “codes for” an organism’s unique traits. In terms of uniqueness and developmental causality, then, genes seem to underlie human identity. Yet with deeper reflection, one might find this commonplace association spurious and misleading: another sacred bovine. Ironically, perhaps, DNA fingerprinting reveals very little about an individual’s DNA, or genome. The technique does not exhaustively profile every form of every gene, as many imagine. Nor does it even sequence the DNA. Rather, it focuses on a rather incidental feature of chromosome structure: differences in noncoding sections of DNA. There, short “nonsense” segments are repeated. The number of repeats, however, varies widely among individuals. Thus, they are convenient markers, or indicators, for identifying a particular organism—or a potential criminal suspect. Each person’s DNA may well be unique, but only a small and physiologically insignificant fragment of it is needed to identify a particular individual. Other biological features function as identifiers, as well. Forensic scientists have long relied on fingerprints and “mug shots,” both introduced into criminology by Charles Darwin’s cousin Francis Galton. They also use hair, skin tone, blood and tissue type, and voice sonograms. Some high-tech security systems—including ones adopted by US immigration—use eye scans. These record the unique pattern of the eye’s iris. (Blood vessel patterns on the retina work as well.) In all these cases, the aim is unambiguous identification. What matters is diagnostically unique properties. So these particular features are effective indicators. At the same time, their functional role is trivial. They are biologically insignificant. They hardly profile someone’s sense of self. Nor do they fully characterize who they are (personally, culturally, or even biologically). Identification and identity are distinct. A unique feature is not necessarily an important feature.
APA, Harvard, Vancouver, ISO, and other styles
9

Cottrell, G. A. "Maximum entropy and plasma physics." In Maximum Entropy in Action, 109–38. Oxford University PressOxford, 1991. http://dx.doi.org/10.1093/oso/9780198539414.003.0005.

Full text
Abstract:
Abstract Experiments on magnetically confined fusion plasmas, for example the Joint European Torus (JET) tokamak, require a range of diagnostic techniques for measurement of plasma quantities such as temperature and density. Usually, some kind of inverse transform is needed to convert measured raw signals into useful data. Also, there are often problems arising from inadequately sampled data and instrumental resolution limitations. It will be explained how the maximum entropy method (MaxEnt) can be applied in such cases with three advantages over conventional analysis: the generation of positive solutions, the suppression of noise and the suppression of spurious transform-related artefacts. MaxEnt is illustrated using a number of examples from the field, involving: the Fourier transform (for example, Michelson interferometry), deconvolution of spectra, and two-dimensional tomography of atomic beams. The advantages of MaxEnt are seen clearly in the problem of extracting the maximum amount of information, for example on plasma profiles, starting from only a limited and noisy data set. Used in this way, the method responds flexibly to the demands of the data and does not suppress any significant features.
APA, Harvard, Vancouver, ISO, and other styles
10

J., Shiny Priyadarshini, and Gladis D. "Analogizing the Thinning Algorithm and Elicitation of Vascular Landmark in Retinal Images." In Ophthalmology, 69–77. IGI Global, 2018. http://dx.doi.org/10.4018/978-1-5225-5195-9.ch005.

Full text
Abstract:
The retinal tissue is composed of network of blood vessels forming a unique biometric pattern. Feature extraction in retinal blood vessel is becoming an emerging trend in the field of personal identification. Because of its unique identity and less vulnerability to noise and distortion it has become one of the most secured biometric identities. The paper highlights the segmentation of blood vessel and the extraction of feature points such as termination and bifurcation points using Zhang Suen's thinning algorithm in retinal images. A comparison has been made and results are analyzed and tabulated between Zhang Suen and Morphological thinning. The count has been taken for both termination and bifurcation markings as spurious and non- spurious minutiae. The spurious minutiae are removed by using the crossing number method. The results clearly depict that the Zhang Suen's thinning algorithm gives better result when compared to morphological thinning.
APA, Harvard, Vancouver, ISO, and other styles

Conference papers on the topic "Spurious features"

1

MaungMaung, AprilPyone, Huy H. Nguyen, Hitoshi Kiya, and Isao Echizen. "Fine-Tuning Text-To-Image Diffusion Models for Class-Wise Spurious Feature Generation." In 2024 IEEE International Conference on Image Processing (ICIP), 3910–16. IEEE, 2024. http://dx.doi.org/10.1109/icip51287.2024.10647627.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Neuhaus, Yannic, Maximilian Augustin, Valentyn Boreiko, and Matthias Hein. "Spurious Features Everywhere - Large-Scale Detection of Harmful Spurious Features in ImageNet." In 2023 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE, 2023. http://dx.doi.org/10.1109/iccv51070.2023.01851.

Full text
APA, Harvard, Vancouver, ISO, and other styles
3

Venkataramani, Rahul, Parag Dutta, Vikram Melapudi, and Ambedkar Dukkipati. "Causal Feature Alignment: Learning to Ignore Spurious Background Features." In 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). IEEE, 2024. http://dx.doi.org/10.1109/wacv57701.2024.00460.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Tornqvist, David, Thomas B. Schon, and Fredrik Gustafsson. "Detecting spurious features using parity space." In 2008 10th International Conference on Control, Automation, Robotics and Vision (ICARCV). IEEE, 2008. http://dx.doi.org/10.1109/icarcv.2008.4795545.

Full text
APA, Harvard, Vancouver, ISO, and other styles
5

Khani, Fereshte, and Percy Liang. "Removing Spurious Features can Hurt Accuracy and Affect Groups Disproportionately." In FAccT '21: 2021 ACM Conference on Fairness, Accountability, and Transparency. New York, NY, USA: ACM, 2021. http://dx.doi.org/10.1145/3442188.3445883.

Full text
APA, Harvard, Vancouver, ISO, and other styles
6

Ramponi, Alan, and Sara Tonelli. "Features or Spurious Artifacts? Data-centric Baselines for Fair and Robust Hate Speech Detection." In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg, PA, USA: Association for Computational Linguistics, 2022. http://dx.doi.org/10.18653/v1/2022.naacl-main.221.

Full text
APA, Harvard, Vancouver, ISO, and other styles
7

Joshi, Nitish, Xiang Pan, and He He. "Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens." In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA, USA: Association for Computational Linguistics, 2022. http://dx.doi.org/10.18653/v1/2022.emnlp-main.666.

Full text
APA, Harvard, Vancouver, ISO, and other styles
8

Du, Yanrui, Jing Yan, Yan Chen, Jing Liu, Sendong Zhao, Qiaoqiao She, Hua Wu, Haifeng Wang, and Bing Qin. "Less Learn Shortcut: Analyzing and Mitigating Learning of Spurious Feature-Label Correlation." In Thirty-Second International Joint Conference on Artificial Intelligence {IJCAI-23}. California: International Joint Conferences on Artificial Intelligence Organization, 2023. http://dx.doi.org/10.24963/ijcai.2023/560.

Full text
Abstract:
Recent research has revealed that deep neural networks often take dataset biases as a shortcut to make decisions rather than understand tasks, leading to failures in real-world applications. In this study, we focus on the spurious correlation between word features and labels that models learn from the biased data distribution of training data. In particular, we define the word highly co-occurring with a specific label as biased word, and the example containing biased word as biased example. Our analysis shows that biased examples are easier for models to learn, while at the time of prediction, biased words make a significantly higher contribution to the models' predictions, and models tend to assign predicted labels over-relying on the spurious correlation between words and labels. To mitigate models' over-reliance on the shortcut (i.e. spurious correlation), we propose a training strategy Less-Learn-Shortcut (LLS): our strategy quantifies the biased degree of the biased examples and down-weights them accordingly. Experimental results on Question Matching, Natural Language Inference and Sentiment Analysis tasks show that LLS is a task-agnostic strategy and can improve the model performance on adversarial data while maintaining good performance on in-domain data.
APA, Harvard, Vancouver, ISO, and other styles
9

Takeda, Keita, Eiji Mitate, and Tomoya Sakai. "Background Subtraction Approach to Unsupervised Cell Segmentation: Toward Excluding Spurious Features in Degraded Cytology Slides." In 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI). IEEE, 2023. http://dx.doi.org/10.1109/isbi53787.2023.10230323.

Full text
APA, Harvard, Vancouver, ISO, and other styles
10

Yadav, Rohan Kumar, Lei Jiao, Ole-Christoffer Granmo, and Morten Goodwin. "Robust Interpretable Text Classification against Spurious Correlations Using AND-rules with Negation." In Thirty-First International Joint Conference on Artificial Intelligence {IJCAI-22}. California: International Joint Conferences on Artificial Intelligence Organization, 2022. http://dx.doi.org/10.24963/ijcai.2022/616.

Full text
Abstract:
The state-of-the-art natural language processing models have raised the bar for excellent performance on a variety of tasks in recent years. However, concerns are rising over their primitive sensitivity to distribution biases that reside in the training and testing data. This issue hugely impacts the performance of the models when exposed to out-of-distribution and counterfactual data. The root cause seems to be that many machine learning models are prone to learn the shortcuts, modelling simple correlations rather than more fundamental and general relationships. As a result, such text classifiers tend to perform poorly when a human makes minor modifications to the data, which raises questions regarding their robustness. In this paper, we employ a rule-based architecture called Tsetlin Machine (TM) that learns both simple and complex correlations by ANDing features and their negations. As such, it generates explainable AND-rules using negated and non-negated reasoning. Here, we explore how non-negated reasoning can be more prone to distribution biases than negated reasoning. We further leverage this finding by adapting the TM architecture to mainly perform negated reasoning using the specificity parameter s. As a result, the AND-rules becomes robust to spurious correlations and can also correctly predict counterfactual data. Our empirical investigation of the model's robustness uses the specificity s to control the degree of negated reasoning. Experiments on publicly available Counterfactually-Augmented Data demonstrate that the negated clauses are robust to spurious correlations and outperform Naive Bayes, SVM, and Bi-LSTM by up to 20 %, and ELMo by almost 6 % on counterfactual test data.
APA, Harvard, Vancouver, ISO, and other styles
You might also be interested in the extended bibliographies on the topic 'Spurious features' for particular source types:
Journal articles
We offer discounts on all premium plans for authors whose works are included in thematic literature selections. Contact us to get a unique promo code!

To the bibliography