Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Visual representation learning.
Статті в журналах з теми "Visual representation learning"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 статей у журналах для дослідження на тему "Visual representation learning".
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
Lee, Jungmin, and Wongyoung Lee. "Aspects of A Study on the Multi Presentational Metaphor Education Using Online Telestration." Korean Society of Culture and Convergence 44, no. 9 (September 30, 2022): 163–73. http://dx.doi.org/10.33645/cnc.2022.9.44.9.163.
Повний текст джерелаАнотація:
This study is an attempt to propose a multiple representational metaphor education model that combines linguistic representation and visual representation using online telestration. The advent of the media and online era has incorporated not only the understanding of linguistic representation l but also the understanding of visual representation into an important phase of cognitive behavior and requires the implementation of online learning. In such an era's needs, it can be said that teaching-learning makes metaphors be used as a tool for thinking and cognition in an online environment, learning leads learners to a new horizon of perception by combining linguistic representation and visual representation. The multiple representational metaphor education model using online telestration will have a two-way dynamic interaction in an online environment, and it will be possible to improve learning capabilities by expressing various representations. Multiple representational metaphor education using online telestration will allow us to consider new perspectives and various possibilities of expression to interpret the world by converging and rephrasing verbal and visual representations using media in an online environment.
2
Yang, Chuanguang, Zhulin An, Linhang Cai, and Yongjun Xu. "Mutual Contrastive Learning for Visual Representation Learning." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 3 (June 28, 2022): 3045–53. http://dx.doi.org/10.1609/aaai.v36i3.20211.
Повний текст джерелаАнотація:
We present a collaborative learning method called Mutual Contrastive Learning (MCL) for general visual representation learning. The core idea of MCL is to perform mutual interaction and transfer of contrastive distributions among a cohort of networks. A crucial component of MCL is Interactive Contrastive Learning (ICL). Compared with vanilla contrastive learning, ICL can aggregate cross-network embedding information and maximize the lower bound to the mutual information between two networks. This enables each network to learn extra contrastive knowledge from others, leading to better feature representations for visual recognition tasks. We emphasize that the resulting MCL is conceptually simple yet empirically powerful. It is a generic framework that can be applied to both supervised and self-supervised representation learning. Experimental results on image classification and transfer learning to object detection show that MCL can lead to consistent performance gains, demonstrating that MCL can guide the network to generate better feature representations. Code is available at https://github.com/winycg/MCL.
3
Khaerun Nisa, Rachmawati, and Reza Muhamad Zaenal. "Analysis Of Students' Mathematical Representation Ability in View of Learning Styles." Indo-MathEdu Intellectuals Journal 4, no. 2 (August 15, 2023): 99–109. http://dx.doi.org/10.54373/imeij.v4i2.119.
Повний текст джерелаАнотація:
Students' mathematical representation ability from the point of view of learning styles, we can see that learning styles play an important role in how students absorb mathematical information. The visual learning style involves using pictures, diagrams and other visual representations to understand math concepts. The auditory learning style involves a preference for listening to oral explanations or participating in discussions. Meanwhile, the kinesthetic learning style involves physical movement to gain an understanding of mathematics. This study aims to determine: 1) the representational ability of students who have a visual learning style, 2) the representational ability of students who have a kinesthetic learning style, 3) the representational ability of students who have an auditory learning style. In this study using descriptive qualitative research methods to determine the ability of mathematical representation which has a learning style for each student in class VII-F SMP Negeri 2 Kuningan. Data were collected through learning style questionnaires, representation ability tests and interviews with students. The results of the learning style questionnaire for the majority of students have a rich visual learning of 42%, kinesthetic 30% and auditory 28%. The results of the representation ability test showed that students with visual learning styles had high mathematical representation abilities, while the majority of students with auditory and kinesthetic learning styles were categorized as moderate. Interviews with students revealed that learning understanding can be understood by means of their learning styles. However, teachers face obstacles in choosing learning models that suit students' learning styles and need to understand individual learning needs. Students' activities in learning mathematics also still need to be improved, especially in the ability to discuss, conclude, and make summaries. Therefore, further efforts are needed to pay attention to differences in student abilities and develop appropriate strategies to increase student interaction and participation in learning mathematics.
4
Kholilatun, Fiki, Nizaruddin Nizaruddin, and F. X. Didik Purwosetiyono. "Kemampuan Representasi Siswa SMP Kelas VIII dalam Menyelesaikan Soal Cerita Materi Peluang Ditinjau dari Gaya Belajar Visual." Jurnal Kualita Pendidikan 4, no. 1 (April 30, 2023): 54–59. http://dx.doi.org/10.51651/jkp.v4i1.339.
Повний текст джерелаАнотація:
This study aims to determine the representational abilities of class VIII junior high school students in solving word problems on opportunity material in terms of student learning styles. This research is a qualitative descriptive study. This research was conducted in class VIII B and VIII E of SMP Miftahul Ulum Boarding School Jogoloyo Demak which were selected based on the results of a questionnaire test. The subjects of this study were selected based on a learning style questionnaire of 60 grade VIII students, 6 students were selected consisting of 2 subjects with a visual learning style, 2 subjects with an auditory learning style, and 2 subjects with a kinesthetic learning style. . The instruments used in this study were a learning style questionnaire test to determine research subjects, math word problems to bring out students' representation abilities, and interview guidelines. The validity of the data uses technical triangulation, namely checking data that has been obtained from the same source using different techniques. The results of the research based on tests and interviews obtained 1) subjects with a visual learning style gave rise to visual representation indicators, and representations of mathematical equations or expressions. 2) subjects with learning styles bring up indicators of visual representation, and representation of equations or mathematical expressions. 3) subjects with kinesthetic learning styles only bring up visual representation indicators.
5
Rif'at, Mohamad, Sudiansyah Sudiansyah, and Khoirunnisa Imama. "Role of visual abilities in mathematics learning: An analysis of conceptual representation." Al-Jabar : Jurnal Pendidikan Matematika 15, no. 1 (June 10, 2024): 87. http://dx.doi.org/10.24042/ajpm.v15i1.22406.
Повний текст джерелаАнотація:
Background: In mathematics education, the understanding of concepts is often influenced by students' visual abilities, making conceptual representation crucial in facilitating comprehension of the material.Aim: This study aims to analyze the role of visual abilities in facilitating the understanding of mathematical concepts through conceptual representation.Method: This research combines quantitative analysis with the use of attribute control diagrams to evaluate data obtained from tasks designed to test students' visual abilities in a mathematical context. These tasks include the manipulation of visual representations and problem-solving using geometric concepts.Results: The findings indicate that 80% of the sample possessed visual abilities that did not meet the expected index, showing a wide variation in students' visual representation abilities. Additionally, most students (70%) were more likely to choose familiar geometric representations in problem-solving, despite difficulties in manipulating more complex concepts.Conclusion: This study demonstrates that students often struggle to effectively utilize visual representations, preferring algebraic approaches that do not fully exploit the potential of conceptual representation. The findings suggest that an increased focus on developing visual abilities, especially in conceptual representation, could strengthen mathematical understanding. Further research is needed to develop intervention strategies that can help students overcome gaps in their visual abilities.
6
Ruliani, Iva Desi, Nizaruddin Nizaruddin, and Yanuar Hery Murtianto. "Profile Analysis of Mathematical Problem Solving Abilities with Krulik & Rudnick Stages Judging from Medium Visual Representation." JIPM (Jurnal Ilmiah Pendidikan Matematika) 7, no. 1 (September 7, 2018): 22. http://dx.doi.org/10.25273/jipm.v7i1.2123.
Повний текст джерелаАнотація:
The ability to solve mathematical problems is very important in learning math and everyday life. According to Krulik & Rudnick there are 5 stages of problem solving that is Read, Explore, Select A Strategy, Solve And Look Back. Mathematical problems require multiple representational skills to communicate problems, one of which is visual representation. Trigonometry is one of the materials that uses visual representation. This research is a qualitative descriptive research that aims to describe the ability of problem solving mathematics with Krulik & Rudnick stages in terms of visual representation. The study was conducted in MAN 2 Brebes. Determination of Subjects in this study using Purposive Sampling. Research instruments used to obtain the required data are visual representation and problem-solving tests, and interview guidelines. The data obtained were analyzed based on the Krulik & Rudnick problem solving indicator. Subjects in this study were subjects with moderate visual representation. Based on the results, problem solving ability of the subject is not fully fulfilled. Subjects with visual representations are able to do problem solving well that is solving the problem through a concept that is understood without visualization of the image. Subjects with visual representations are having a schematic visual representation type.
7
Zha, B., and A. Yilmaz. "LEARNING MAPS FOR OBJECT LOCALIZATION USING VISUAL-INERTIAL ODOMETRY." ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences V-1-2020 (August 3, 2020): 343–50. http://dx.doi.org/10.5194/isprs-annals-v-1-2020-343-2020.
Повний текст джерелаАнотація:
Abstract. Objects follow designated path on maps, such as vehicles travelling on a road. This observation signifies topological representation of objects’ motion on the map. Considering the position of object is unknown initially, as it traverses the map by moving and turning, the spatial uncertainty of its whereabouts reduces to a single location as the motion trajectory would fit only to a certain map trajectory. Inspired by this observation, we propose a novel end-to-end localization approach based on topological maps that exploits the object motion and learning the map using an recurrent neural network (RNN) model. The core of the proposed method is to learn potential motion patterns from the map and perform trajectory classification in the map’s edge-space. Two different trajectory representations, namely angle representation and augmented angle representation (incorporates distance traversed) are considered and an RNN is trained from the map for each representation to compare their performances. The localization accuracy in the tested map for the angle and augmented angle representations are 90.43% and 96.22% respectively. The results from the actual visual-inertial odometry have shown that the proposed approach is able to learn the map and localize objects based on their motion.
8
Moghaddam, B., and A. Pentland. "Probabilistic visual learning for object representation." IEEE Transactions on Pattern Analysis and Machine Intelligence 19, no. 7 (July 1997): 696–710. http://dx.doi.org/10.1109/34.598227.
Повний текст джерела
9
He, Xiangteng, and Yuxin Peng. "Fine-Grained Visual-Textual Representation Learning." IEEE Transactions on Circuits and Systems for Video Technology 30, no. 2 (February 2020): 520–31. http://dx.doi.org/10.1109/tcsvt.2019.2892802.
Повний текст джерела
10
Liu, Qiyuan, Qi Zhou, Rui Yang, and Jie Wang. "Robust Representation Learning by Clustering with Bisimulation Metrics for Visual Reinforcement Learning with Distractions." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 7 (June 26, 2023): 8843–51. http://dx.doi.org/10.1609/aaai.v37i7.26063.
Повний текст джерелаАнотація:
Recent work has shown that representation learning plays a critical role in sample-efficient reinforcement learning (RL) from pixels. Unfortunately, in real-world scenarios, representation learning is usually fragile to task-irrelevant distractions such as variations in background or viewpoint. To tackle this problem, we propose a novel clustering-based approach, namely Clustering with Bisimulation Metrics (CBM), which learns robust representations by grouping visual observations in the latent space. Specifically, CBM alternates between two steps: (1) grouping observations by measuring their bisimulation distances to the learned prototypes; (2) learning a set of prototypes according to the current cluster assignments. Computing cluster assignments with bisimulation metrics enables CBM to capture task-relevant information, as bisimulation metrics quantify the behavioral similarity between observations. Moreover, CBM encourages the consistency of representations within each group, which facilitates filtering out task-irrelevant information and thus induces robust representations against distractions. An appealing feature is that CBM can achieve sample-efficient representation learning even if multiple distractions exist simultaneously. Experiments demonstrate that CBM significantly improves the sample efficiency of popular visual RL algorithms and achieves state-of-the-art performance on both multiple and single distraction settings. The code is available at https://github.com/MIRALab-USTC/RL-CBM.
11
Erita, Selvia, Tami Mulyani, and Aan Putra. "Analysis Of Mathematic Representation Ability In Online Learning." Mathline : Jurnal Matematika dan Pendidikan Matematika 8, no. 1 (February 19, 2023): 101–12. http://dx.doi.org/10.31943/mathline.v8i1.259.
Повний текст джерелаАнотація:
ABSTRACT The outbreak of the corona virus has had an impact on various sectors, especially the education sector. The application of online learning has been implemented in almost all schools in Indonesia. However, judging from the results of various studies and some experiences, the online learning process in Indonesia has previously been carried out well. Mathematical representation skills are needed so that students understand mathematical concepts well. This study aims to analyze the ability of mathematical representation in online learning. The method used is descriptive qualitative type. The instrument used is a mathematical representation ability test and interview guidelines. The data analysis technique used is the stages of data reduction, data display and making overall conclusions. Data analysis of test results was used to determine the level of students' mathematical representation abilities, while interviews were to strengthen test results and determine student learning constraints. Based on the results of the study, it was found that students' visual representations were more dominant than other types of representations. The students' mathematical visual representation ability is in the sufficient (moderate) category, while the students' symbolic representation ability is very poor and the students' verbal representation ability is in the poor category.
12
Cahyaningrum, Ira Yoshita, and Surya Sari Faradiba. "DECISION MAKING IN POLICY LEARNING MODEL THROUGH STUDENT MATHEMATICS REPRESENTATION ABILITY TEST IN LEARNING TWO VARIABLE LINEAR EQUATIONS." dia 20, no. 01 (May 24, 2022): 343–53. http://dx.doi.org/10.30996/dia.v20i01.6435.
Повний текст джерелаАнотація:
This study aims to make decisions about learning model policies by exploring students' mathematical representations in learning mathematics. The benefit of the ability to represent mathematics is that it facilitates problem-solving. Students need to train and develop things related to mathematical representation in a lesson. The form of mathematical representation used is a visual representation, making mathematical models, and using words in problem-solving. From the results of the study, it was found that the students' mathematical representation skills were linear equations of two variables, namely students as the first subject in the very capable category of students fulfilling complete and very good mathematical representations. Students with moderate academic ability in the second subject and meeting the mathematical representation ability were quite good, while students in the third subject with existing indicators had incomplete results and did not meet the mathematical representation.
13
Adi, Yogi Kuncoro, Ari Widodo, Wahyu Sopandi, and Muslim. "Students’ Visual Representation of Lights and Visions." Jurnal Penelitian Pendidikan IPA 9, no. 10 (October 25, 2023): 8546–53. http://dx.doi.org/10.29303/jppipa.v9i10.5240.
Повний текст джерелаАнотація:
Visual representations are used by scientists to communicate scientific conceptions and are used by teachers to teach science in the classroom. The 2013 Curriculum textbook reduces this visual representation. Meanwhile, visual representations will help students develop a comprehensive understanding of the concept. This case study research aims to reveal cases of misconceptions in the visual representation of students at an X elementary school. We used observations of fifty-nine fourth-grade elementary school students to find students with different cases of misconceptions. Eleven students were further identified using interviews and drawing tests. We analysed the data qualitatively based on the collection of these two types of data. We found misconceptions in the representation of luminous objects and how students draw visions of luminous objects and non-luminous objects. Research results showed that we found cases of misconceptions similar to the findings from previous studies. While light and vision are prerequisite concepts, a student's conception of vision is affected when he has a misconception about light. Content can be developed by paying attention to the various modes of representation, conceptual change, and learning progression in the future. The pattern of learning progression can be studied in more detail using the microgenetic method.
14
Nai, Ruiqian, Zixin Wen, Ji Li, Yuanzhi Li, and Yang Gao. "Revisiting Disentanglement in Downstream Tasks: A Study on Its Necessity for Abstract Visual Reasoning." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 13 (March 24, 2024): 14405–13. http://dx.doi.org/10.1609/aaai.v38i13.29354.
Повний текст джерелаАнотація:
In representation learning, a disentangled representation is highly desirable as it encodes generative factors of data in a separable and compact pattern. Researchers have advocated leveraging disentangled representations to complete downstream tasks with encouraging empirical evidence. This paper further investigates the necessity of disentangled representation in downstream applications. Specifically, we show that dimension-wise disentangled representations are unnecessary on a fundamental downstream task, abstract visual reasoning. We provide extensive empirical evidence against the necessity of disentanglement, covering multiple datasets, representation learning methods, and downstream network architectures. Furthermore, our findings suggest that the informativeness of representations is a better indicator of downstream performance than disentanglement. Finally, the positive correlation between informativeness and disentanglement explains the claimed usefulness of disentangled representations in previous works. The source code is available at https://github.com/Richard-coder-Nai/disentanglement-lib-necessity.git
15
Zhang, Jingran, Xing Xu, Fumin Shen, Huimin Lu, Xin Liu, and Heng Tao Shen. "Enhancing Audio-Visual Association with Self-Supervised Curriculum Learning." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 4 (May 18, 2021): 3351–59. http://dx.doi.org/10.1609/aaai.v35i4.16447.
Повний текст джерелаАнотація:
The recent success of audio-visual representations learning can be largely attributed to their pervasive concurrency property, which can be used as a self-supervision signal and extract correlation information. While most recent works focus on capturing the shared associations between the audio and visual modalities, they rarely consider multiple audio and video pairs at once and pay little attention to exploiting the valuable information within each modality. To tackle this problem, we propose a novel audio-visual representation learning method dubbed self-supervised curriculum learning (SSCL) under the teacher-student learning manner. Specifically, taking advantage of contrastive learning, a two-stage scheme is exploited, which transfers the cross-modal information between teacher and student model as a phased process. The proposed SSCL approach regards the pervasive property of audiovisual concurrency as latent supervision and mutually distills the structure knowledge of visual to audio data. Notably, the SSCL method can learn discriminative audio and visual representations for various downstream applications. Extensive experiments conducted on both action video recognition and audio sound recognition tasks show the remarkably improved performance of the SSCL method compared with the state-of-the-art self-supervised audio-visual representation learning methods.
16
Yuan, Hangjie, and Dong Ni. "Learning Visual Context for Group Activity Recognition." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 4 (May 18, 2021): 3261–69. http://dx.doi.org/10.1609/aaai.v35i4.16437.
Повний текст джерелаАнотація:
Group activity recognition aims to recognize an overall activity in a multi-person scene. Previous methods strive to reason on individual features. However, they under-explore the person-specific contextual information, which is significant and informative in computer vision tasks. In this paper, we propose a new reasoning paradigm to incorporate global contextual information. Specifically, we propose two modules to bridge the gap between group activity and visual context. The first is Transformer based Context Encoding (TCE) module, which enhances individual representation by encoding global contextual information to individual features and refining the aggregated information. The second is Spatial-Temporal Bilinear Pooling (STBiP) module. It firstly further explores pairwise relationships for the context encoded individual representation, then generates semantic representations via gated message passing on a constructed spatial-temporal graph. On their basis, we further design a two-branch model that integrates the designed modules into a pipeline. Systematic experiments demonstrate each module's effectiveness on either branch. Visualizations indicate that visual contextual cues can be aggregated globally by TCE. Moreover, our method achieves state-of-the-art results on two widely used benchmarks using only RGB images as input and 2D backbones.
17
Hemavathy, J., E. Arul Jothi, R. Nishalini, and M. Oviya. "Maneuvers of Multi Perspective Media Retrieval." International Journal of Research in Engineering, Science and Management 3, no. 9 (September 17, 2020): 71–74. http://dx.doi.org/10.47607/ijresm.2020.290.
Повний текст джерелаАнотація:
Recently, Learning Machines have achieved a measure of success in the representation of multiple views. Since the effectiveness of data mining methods is highly dependent on the ability to produce data representation, learning multi-visual representation has become a very promising topic with widespread use. It is an emerging data mining guide that looks at multidisciplinary learning to improve overall performance. Multi-view reading is also known as data integration or data integration from multiple feature sets. In general, learning the representation of multiple views is able to learn the informative and cohesive representation that leads to the improvement in the performance of predictors. Therefore, learning multi-view representation has been widely used in many real-world applications including media retrieval, native language processing, video analysis, and a recommendation program. We propose two main stages of learning multidisciplinary representation: (i) alignment of multidisciplinary representation, which aims to capture relationships between different perspectives on content alignment; (ii) a combination of different visual representations, which seeks to combine different aspects learned from many different perspectives into a single integrated representation. Both of these strategies seek to use the relevant information contained in most views to represent the data as a whole. In this project we use the concept of canonical integration analysis to get more details. Encouraged by the success of in-depth reading, in-depth reading representation of multiple theories has attracted a lot of attention in media access due to its ability to read explicit visual representation.
18
Basimin, Mudzuna Quraisyah, Habiddin Habiddin, and Ridwan Joharmawan. "Higher Order Thinking Skills and Visual Representations of Chemical Concepts: A Literature Review." Hydrogen: Jurnal Kependidikan Kimia 11, no. 6 (December 30, 2023): 1043. http://dx.doi.org/10.33394/hjkk.v11i6.10173.
Повний текст джерелаАнотація:
Efforts to assist students in understanding generally abstract chemical concepts are widely done using visual representations as a form of multiple representations in chemistry. This article evaluates and identifies articles from the year (2013-2023) through search engines that provide international services and national journal pages that can be accessed using 4 databases, namely, science direct, eric, google scholar, and crossref. Based on predefined criteria for the use of visual representation in chemistry to improve Higher Order Thinking Skills, 13 relevant articles were obtained. The results of the review show that visual representation can be utilized to train and improve higher-order thinking skills, especially critical, logical, reflective, metacognitive, and creative thinking. Visual representation has also been applied to several approaches or learning models such as Multiple Representation, Particulate Representation, 5R, SWH, Marzano's Taxonomy, Use of Concept Maps, and PcBL.
19
Wu, Hui, Min Wang, Wengang Zhou, Yang Hu, and Houqiang Li. "Learning Token-Based Representation for Image Retrieval." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 3 (June 28, 2022): 2703–11. http://dx.doi.org/10.1609/aaai.v36i3.20173.
Повний текст джерелаАнотація:
In image retrieval, deep local features learned in a data-driven manner have been demonstrated effective to improve retrieval performance. To realize efficient retrieval on large image database, some approaches quantize deep local features with a large codebook and match images with aggregated match kernel. However, the complexity of these approaches is non-trivial with large memory footprint, which limits their capability to jointly perform feature learning and aggregation. To generate compact global representations while maintaining regional matching capability, we propose a unified framework to jointly learn local feature representation and aggregation. In our framework, we first extract local features using CNNs. Then, we design a tokenizer module to aggregate them into a few visual tokens, each corresponding to a specific visual pattern. This helps to remove background noise, and capture more discriminative regions in the image. Next, a refinement block is introduced to enhance the visual tokens with self-attention and cross-attention. Finally, different visual tokens are concatenated to generate a compact global representation. The whole framework is trained end-to-end with image-level labels. Extensive experiments are conducted to evaluate our approach, which outperforms the state-of-the-art methods on the Revisited Oxford and Paris datasets.
20
Ramdani, Rida, and Arip Nurahman. "Exploring senior high school students’ preconceptions of collision concepts using visual representation." Research in Physics Education 2, no. 2 (December 26, 2023): 59–68. http://dx.doi.org/10.31980/ripe.v2i2.31.
Повний текст джерелаАнотація:
Understanding students’ preconceptions is an important initial step towards meaningful and appropriate physics learning. Particularly in the concept of collisions, students’ understanding of collision concepts often does not align with those established by scientists. Therefore, to explore students’ preconceptions, visual representation enables students to connect their experiences with specific concepts. This aids in revealing and diagnosing students’ preconceptions. This research aims to analyze high school students’ preconceptions regarding collision concepts by examining their visual representations. The research design employs a qualitative descriptive research design with data collection methods including observation sheets and interviews. Observation sheets containing visual representations by students are analyzed using a four-step semiotic analysis approach, while interview data is analyzed thematically. Seventy-two tenth-grade students were sampled offline using convenience sampling. The research findings reveal that students’ use of visual representations indicates diverse preconceptions about collision concepts, categorizing them at the macroscopic level. This data is examined across four categories of student visuals: (1) 11 visuals that are correct; (2) 18 visuals based on the objects used; (3) 34 visuals depicting different types of collisions; (4) and 29 visuals that are incorrect.
21
Syafitri, Aulia, Roseli Theis, and Dewi Iriani. "ANALISIS KESULITAN KEMAMPUAN REPRESENTASI MATEMATIS SISWA EKSTROVERT DALAM MENYELESAIKAN SOAL MATEMATIKA PADA MATERI ALJABAR." Absis: Mathematics Education Journal 3, no. 1 (March 20, 2021): 16. http://dx.doi.org/10.32585/absis.v3i1.1382.
Повний текст джерелаАнотація:
The objective of this study is to analyze the difficulty of representation ability Mathematical extrovert students in class VII-D in solving math problems on algebraic material, as well as knowing the factors that cause difficulties experienced by students in meet the indicators of mathematical representation ability. The ability of mathematical representations is measured based on Its aspects include visual representation aspects, expression representation aspects mathematics, and aspects of the representation of words or written text. This is qualitative research using a descriptive approach. This is carried out at SMP Negeri 22 Jambi with 4 students from class VII-D. The results showed that SE1, SE2, SE3 and SE4 are students with extrovert personalities do not have difficulty in the visual aspect. On aspects of representation of mathematical equations or expressions of Students SE1, SE3, and SE4 having difficulty making mathematical models or equations. On that aspect word or written text students SE2, SE3, and SE4 have difficulty when determine what steps will be taken to solve the problem mathematical. Factors causing difficulties in the mathematical representation ability are on visual aspects, representational aspects of mathematical expressions, and aspects of word representation or written text is a non-cognitive learning factor
22
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.
23
Yue, Yang, Bingyi Kang, Zhongwen Xu, Gao Huang, and Shuicheng Yan. "Value-Consistent Representation Learning for Data-Efficient Reinforcement Learning." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 9 (June 26, 2023): 11069–77. http://dx.doi.org/10.1609/aaai.v37i9.26311.
Повний текст джерелаАнотація:
Deep reinforcement learning (RL) algorithms suffer severe performance degradation when the interaction data is scarce, which limits their real-world application. Recently, visual representation learning has been shown to be effective and promising for boosting sample efficiency in RL. These methods usually rely on contrastive learning and data augmentation to train a transition model, which is different from how the model is used in RL---performing value-based planning. Accordingly, the learned representation by these visual methods may be good for recognition but not optimal for estimating state value and solving the decision problem. To address this issue, we propose a novel method, called value-consistent representation learning (VCR), to learn representations that are directly related to decision-making. More specifically, VCR trains a model to predict the future state (also referred to as the "imagined state'') based on the current one and a sequence of actions. Instead of aligning this imagined state with a real state returned by the environment, VCR applies a Q value head on both of the states and obtains two distributions of action values. Then a distance is computed and minimized to force the imagined state to produce a similar action value prediction as that by the real state. We develop two implementations of the above idea for the discrete and continuous action spaces respectively. We conduct experiments on Atari 100k and DeepMind Control Suite benchmarks to validate their effectiveness for improving sample efficiency. It has been demonstrated that our methods achieve new state-of-the-art performance for search-free RL algorithms.
24
Keller, Patrick, Abdoul Kader Kaboré, Laura Plein, Jacques Klein, Yves Le Traon, and Tegawendé F. Bissyandé. "What You See is What it Means! Semantic Representation Learning of Code based on Visualization and Transfer Learning." ACM Transactions on Software Engineering and Methodology 31, no. 2 (April 30, 2022): 1–34. http://dx.doi.org/10.1145/3485135.
Повний текст джерелаАнотація:
Recent successes in training word embeddings for Natural Language Processing ( NLP ) tasks have encouraged a wave of research on representation learning for source code, which builds on similar NLP methods. The overall objective is then to produce code embeddings that capture the maximum of program semantics. State-of-the-art approaches invariably rely on a syntactic representation (i.e., raw lexical tokens, abstract syntax trees, or intermediate representation tokens) to generate embeddings, which are criticized in the literature as non-robust or non-generalizable. In this work, we investigate a novel embedding approach based on the intuition that source code has visual patterns of semantics. We further use these patterns to address the outstanding challenge of identifying semantic code clones. We propose the WySiWiM ( ‘ ‘What You See Is What It Means ” ) approach where visual representations of source code are fed into powerful pre-trained image classification neural networks from the field of computer vision to benefit from the practical advantages of transfer learning. We evaluate the proposed embedding approach on the task of vulnerable code prediction in source code and on two variations of the task of semantic code clone identification: code clone detection (a binary classification problem), and code classification (a multi-classification problem). We show with experiments on the BigCloneBench (Java), Open Judge (C) that although simple, our WySiWiM approach performs as effectively as state-of-the-art approaches such as ASTNN or TBCNN. We also showed with data from NVD and SARD that WySiWiM representation can be used to learn a vulnerable code detector with reasonable performance (accuracy ∼90%). We further explore the influence of different steps in our approach, such as the choice of visual representations or the classification algorithm, to eventually discuss the promises and limitations of this research direction.
25
Maduratna, Melinia, and Ummi Nur Afinni Dwi Jayanti. "Visual Representation of Biology Books on Circulatory System Material." BIOEDUSCIENCE 6, no. 2 (August 31, 2022): 124–36. http://dx.doi.org/10.22236/j.bes/629415.
Повний текст джерелаАнотація:
Background: Representation is a tool to improve students' communication, interpreting, and problem-solving skills. Visual representation (RV) can provide information about understanding the object/phenomenon observed about the concept under study. This study aimed to determine the relationship between visual representations and material content and the relationship between visual representations and the reality of images contained in the Circulation System material. Methods: This study used a content analysis method with a qualitative data analysis technique. Results: The analysis results show a relationship compared to the category of significant relationships on the circulation system's material and the relationship of symbolic visual representations. Conclusions: The analysis that has been carried out on the visual representation of SMA/MA biology books in the city of Stabat for the circulation system material, it can be concluded that the Visual Representation relationship contained in SMA/MA biology books was found to be more dominant for the category of no relationship compared to the category of a significant relationship in circulation system material presented. In addition, when analyzed concerning the reality of the image, the visual representation displays the relationship of symbolic visual representations in the analyzed biology textbooks. In addition to choosing a textbook, the teacher should pay attention not only to the content of the learning material but also to look at the components in the book, one of which is the picture in the textbook. This will later be useful to support students' understanding of the concepts being taught.
26
Suryani, Fifi, and Mashuri Mashuri. "Students’ Mathematical Representation Ability in Cooperative Learning Type of Reciprocal Peer Tutoring from Learning Style." Unnes Journal of Mathematics Education 12, no. 1 (March 31, 2023): 13–22. http://dx.doi.org/10.15294/ujme.v12i1.67545.
Повний текст джерелаАнотація:
This study aims to analyze whether the Cooperative Learning type of Reciprocal Peer Tutoring (RPT) is effective in enhancing students' mathematical representation abilities, whether it is more effective than PBL in enhancing students' mathematical representation abilities, how learning styles influence students' mathematical representation abilities in Cooperative Learning type of RPT, and to describe the ability of mathematical representation in terms of learning styles in Cooperative Learning type of RPT. The research method used was a mixed method, and the design used was sequential explanatory. The sampling technique used was random sampling by class. The results showed that: (1) the ability of mathematical representation in the Cooperative Learning type of RPT achieved classical completeness; (2) the average mathematical representation ability in the Cooperative Learning type of RPT was higher than that of PBL; (3) the proportion of completeness of mathematical representation ability in Cooperative Learning type of RPT was higher than that of the PBL class, indicating that Cooperative Learning type of RPT is more effective in enhancing mathematical representation abilities; and (4) there is a positive influence of learning styles on students' mathematical representation ability in Cooperative Learning type of RPT. Subjects with visual learning styles are able to fulfill visual and verbal indicators and tend to be able to fulfill symbolic indicators. Subjects with auditory learning styles tend to be able to fulfill visual and symbolic indicators but tend to be less able to fulfill verbal indicators. Subjects with kinaesthetic learning styles are able to meet visual indicators.
27
Joo, Hyun, Jongchan Park, and Dongsik Kim. "Visual representation fidelity and self‐explanation prompts in multi‐representational adaptive learning." Journal of Computer Assisted Learning 37, no. 4 (April 6, 2021): 1091–106. http://dx.doi.org/10.1111/jcal.12548.
Повний текст джерела
28
Proboretno, Setyaning, and Pradnyo Wijayanti. "Representasi Matematis Siswa SMP dalam Meyelesaikan Masalah Segiempat Ditinjau dari Perbedaan Jenis Kelamin." MATHEdunesa 8, no. 3 (August 12, 2019): 472–76. http://dx.doi.org/10.26740/mathedunesa.v8n3.p472-476.
Повний текст джерелаАнотація:
Mathematical representation has an important role to help students understand and solve quadrilateral problems in mathematics learning. Students will use different forms of mathematical representation to solve a quadrilateral problem. This allows that the form of mathematical representation used by male and female students is different. The purpose of this study was to describe the mathematical representation of male and female junior high school students in solving quadrilateral problems. This research is classified into descriptive qualitative research using test and interview methods. The results of this study indicate that male students use visual-spatial representations in the form of images to represent an object that is in the problem solving test. In addition, they use visual-spatial representations and formal-notational representations to reveal information about a problem. During the problem solving process, dominant male students use formal-notational representation. They also explained verbally each step of the completion in detail and in order. Dominant female students use formal-notational representation to write information and solve a problem. To represent an object in a problem solving test, they use visual-spatial representations. Female students also use verbal representations to explain each step of solving problems.Keywords: mathematical representation, quadrilateral problems, gender
29
Kang, Kyuchang, and Changseok Bae. "Memory Model for Morphological Semantics of Visual Stimuli Using Sparse Distributed Representation." Applied Sciences 11, no. 22 (November 15, 2021): 10786. http://dx.doi.org/10.3390/app112210786.
Повний текст джерелаАнотація:
Recent achievements on CNN (convolutional neural networks) and DNN (deep neural networks) researches provide a lot of practical applications on computer vision area. However, these approaches require construction of huge size of training data for learning process. This paper tries to find a way for continual learning which does not require prior high-cost training data construction by imitating a biological memory model. We employ SDR (sparse distributed representation) for information processing and semantic memory model, which is known as a representation model of firing patterns on neurons in neocortex area. This paper proposes a novel memory model to reflect remembrance of morphological semantics of visual input stimuli. The proposed memory model considers both memory process and recall process separately. First, memory process converts input visual stimuli to sparse distributed representation, and in this process, morphological semantic of input visual stimuli can be preserved. Next, recall process can be considered by comparing sparse distributed representation of new input visual stimulus and remembered sparse distributed representations. Superposition of sparse distributed representation is used to measure similarities. Experimental results using 10,000 images in MNIST (Modified National Institute of Standards and Technology) and Fashion-MNIST data sets show that the sparse distributed representation of the proposed model efficiently keeps morphological semantic of the input visual stimuli.
30
Folstein, Jonathan R., Shamsi S. Monfared, and Trevor Maravel. "The effect of category learning on visual attention and visual representation." Psychophysiology 54, no. 12 (August 4, 2017): 1855–71. http://dx.doi.org/10.1111/psyp.12966.
Повний текст джерела
31
Aslamiah, Suaibatul, Rahmah Johar, and Erni Maidiyah. "Kemampuan Representasi Matematis Siswa melalui Model Problem Based Learning pada Materi Lingkaran dengan Konteks Kepramukaan di SMP." JURNAL EKSAKTA PENDIDIKAN (JEP) 3, no. 2 (November 29, 2019): 92. http://dx.doi.org/10.24036/jep/vol3-iss2/336.
Повний текст джерелаАнотація:
Mathematical representation ability is a useful ability for students to develop and optimize their mathematical abilities. The real situation shows that the importance of the ability of mathematical representation has not been matched by student achievement. This condition can be solved by applying problem based learning model to learning in a context that is familiar to students such as scouts. The aim of this research is to explain the ability of students’ mathematical representation through the Problem Based Learning model with scouts context on circle material at SMPN 8 Banda Aceh. This research used qualitative approach with a type of research is descriptive. Data obtained through tests of student’ mathematical representation abilities and interviews. The ability of students’ mathematical representations was analyzed by utilizing the data obtained in student learning in groups through PBL model and data of result test after the application of PBL model. The achievement of this research is the ability of students’ mathematical representation during learning with PBL model in groups is accordance with the ability of students’ mathematical representation after applying the PBL model. The mathematical representation ability of high ability students is able to complete all of the indicators of the mathematical representation ability, students who are in the medium ability only complete indicators of visual and symbolic representation, and low ability students only able to complete indicator of visual representation, although not perfectly.
32
Liu, Sheng, Kevin Lin, Lijuan Wang, Junsong Yuan, and Zicheng Liu. "OVIS: Open-Vocabulary Visual Instance Search via Visual-Semantic Aligned Representation Learning." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 2 (June 28, 2022): 1773–81. http://dx.doi.org/10.1609/aaai.v36i2.20070.
Повний текст джерелаАнотація:
We introduce the task of open-vocabulary visual instance search (OVIS). Given an arbitrary textual search query, Open-vocabulary Visual Instance Search (OVIS) aims to return a ranked list of visual instances, i.e., image patches, that satisfies the search intent from an image database. The term ``open vocabulary'' means that there are neither restrictions to the visual instance to be searched nor restrictions to the word that can be used to compose the textual search query. We propose to address such a search challenge via visual-semantic aligned representation learning (ViSA). ViSA leverages massive image-caption pairs as weak image-level (not instance-level) supervision to learn a rich cross-modal semantic space where the representations of visual instances (not images) and those of textual queries are aligned, thus allowing us to measure the similarities between any visual instance and an arbitrary textual query. To evaluate the performance of ViSA, we build two datasets named OVIS40 and OVIS1600 and also introduce a pipeline for error analysis. Through extensive experiments on the two datasets, we demonstrate ViSA's ability to search for visual instances in images not available during training given a wide range of textual queries including those composed of uncommon words. Experimental results show that ViSA achieves an mAP@50 of 27.8% on OVIS40 and achieves a recall@30 of 21.3% on OVIS1400 dataset under the most challenging settings.
33
Amalia, Aula, Nurina Happy, and FX Didik Purwosetiyono. "Profil Kemampuan Representasi Siswa Dalam Memecahkan Masalah Matematika Ditinjau Dari Gaya Belajar." Phenomenon : Jurnal Pendidikan MIPA 11, no. 1 (December 26, 2021): 15–28. http://dx.doi.org/10.21580/phen.2021.11.1.6521.
Повний текст джерелаАнотація:
This study aims to determaine the profile of the representation ability of juniorhigh school students in terms of learning styles. This type of research wasdescriptive qualitative research. The subjects taken were three junior highschool students of eight grade, each of whom had a visual learning style, andauditory learning style, and kinesthetic learning style. The data was collectedusing a learning style scale, written tests, interviews and documentation. Thedata analysis technique was carried out in 3 stages, reduction, datapresentation, and drawing conclucions or verification. The validity of the dataused time triangulation, comparaing the results of the representation abilitytest with the results of interviews in the first and second stages. The analysiswas developed based on indicators of representational ability by taking inroaccount student learning styles. Based on the results of the analysis, it isknown that subjects with visual, auditory and kinesthetic learning styles havelow verbal representation abilities.
34
Zhao, Qilu, and Junyu Dong. "Self-supervised representation learning by predicting visual permutations." Knowledge-Based Systems 210 (December 2020): 106534. http://dx.doi.org/10.1016/j.knosys.2020.106534.
Повний текст джерела
35
Zhang, Hong, Wenping Zhang, Wenhe Liu, Xin Xu, and Hehe Fan. "Multiple kernel visual-auditory representation learning for retrieval." Multimedia Tools and Applications 75, no. 15 (February 23, 2016): 9169–84. http://dx.doi.org/10.1007/s11042-016-3294-5.
Повний текст джерела
36
Peng, Andrew W., Jiangpeng He, and Fengqing Zhu. "Self-supervised visual representation learning on food images." Electronic Imaging 35, no. 7 (January 16, 2023): 269–1. http://dx.doi.org/10.2352/ei.2023.35.7.image-269.
Повний текст джерела
37
Jean, Neal, Sherrie Wang, Anshul Samar, George Azzari, David Lobell, and Stefano Ermon. "Tile2Vec: Unsupervised Representation Learning for Spatially Distributed Data." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 3967–74. http://dx.doi.org/10.1609/aaai.v33i01.33013967.
Повний текст джерелаАнотація:
Geospatial analysis lacks methods like the word vector representations and pre-trained networks that significantly boost performance across a wide range of natural language and computer vision tasks. To fill this gap, we introduce Tile2Vec, an unsupervised representation learning algorithm that extends the distributional hypothesis from natural language — words appearing in similar contexts tend to have similar meanings — to spatially distributed data. We demonstrate empirically that Tile2Vec learns semantically meaningful representations for both image and non-image datasets. Our learned representations significantly improve performance in downstream classification tasks and, similarly to word vectors, allow visual analogies to be obtained via simple arithmetic in the latent space.
38
Kuo, Yen-Ling. "Learning Representations for Robust Human-Robot Interaction." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 20 (March 24, 2024): 22673. http://dx.doi.org/10.1609/aaai.v38i20.30289.
Повний текст джерелаАнотація:
For robots to robustly and flexibly interact with humans, they need to acquire skills to use across scenarios. One way to enable the generalization of skills is to learn representations that are useful for downstream tasks. Learning a representation for interactions requires an understanding of what (e.g., objects) as well as how (e.g., actions, controls, and manners) to interact with. However, most existing language or visual representations mainly focus on objects. To enable robust human-robot interactions, we need a representation that is not just grounded at the object level but to reason at the action level. The ability to reason about an agent’s own actions and other’s actions will be crucial for long-tail interactions. My research focuses on leveraging the compositional nature of language and reward functions to learn representations that generalize to novel scenarios. Together with the information from multiple modalities, the learned representation can reason about task progress, future behaviors, and the goals/beliefs of an agent. The above ideas have been demonstrated in my research on building robots to understand language and engage in social interactions.
39
Liu, Yongfei, Bo Wan, Xiaodan Zhu, and Xuming He. "Learning Cross-Modal Context Graph for Visual Grounding." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 07 (April 3, 2020): 11645–52. http://dx.doi.org/10.1609/aaai.v34i07.6833.
Повний текст джерелаАнотація:
Visual grounding is a ubiquitous building block in many vision-language tasks and yet remains challenging due to large variations in visual and linguistic features of grounding entities, strong context effect and the resulting semantic ambiguities. Prior works typically focus on learning representations of individual phrases with limited context information. To address their limitations, this paper proposes a language-guided graph representation to capture the global context of grounding entities and their relations, and develop a cross-modal graph matching strategy for the multiple-phrase visual grounding task. In particular, we introduce a modular graph neural network to compute context-aware representations of phrases and object proposals respectively via message propagation, followed by a graph-based matching module to generate globally consistent localization of grounding phrases. We train the entire graph neural network jointly in a two-stage strategy and evaluate it on the Flickr30K Entities benchmark. Extensive experiments show that our method outperforms the prior state of the arts by a sizable margin, evidencing the efficacy of our grounding framework. Code is available at https://github.com/youngfly11/LCMCG-PyTorch.
40
Yang, Fan, Wei Li, Menglong Yang, Binbin Liang, and Jianwei Zhang. "Multi-Modal Disordered Representation Learning Network for Description-Based Person Search." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 15 (March 24, 2024): 16316–24. http://dx.doi.org/10.1609/aaai.v38i15.29567.
Повний текст джерелаАнотація:
Description-based person search aims to retrieve images of the target identity via textual descriptions. One of the challenges for this task is to extract discriminative representation from images and descriptions. Most existing methods apply the part-based split method or external models to explore the fine-grained details of local features, which ignore the global relationship between partial information and cause network instability. To overcome these issues, we propose a Multi-modal Disordered Representation Learning Network (MDRL) for description-based person search to fully extract the visual and textual representations. Specifically, we design a Cross-modality Global Feature Learning Architecture to learn the global features from the two modalities and meet the demand of the task. Based on our global network, we introduce a Disorder Local Learning Module to explore local features by a disordered reorganization strategy from both visual and textual aspects and enhance the robustness of the whole network. Besides, we introduce a Cross-modality Interaction Module to guide the two streams to extract visual or textual representations considering the correlation between modalities. Extensive experiments are conducted on two public datasets, and the results show that our method outperforms the state-of-the-art methods on CUHK-PEDES and ICFG-PEDES datasets and achieves superior performance.
41
She, Dong-Yu, and Kun Xu. "Contrastive Self-supervised Representation Learning Using Synthetic Data." International Journal of Automation and Computing 18, no. 4 (May 11, 2021): 556–67. http://dx.doi.org/10.1007/s11633-021-1297-9.
Повний текст джерелаАнотація:
AbstractLearning discriminative representations with deep neural networks often relies on massive labeled data, which is expensive and difficult to obtain in many real scenarios. As an alternative, self-supervised learning that leverages input itself as supervision is strongly preferred for its soaring performance on visual representation learning. This paper introduces a contrastive self-supervised framework for learning generalizable representations on the synthetic data that can be obtained easily with complete controllability. Specifically, we propose to optimize a contrastive learning task and a physical property prediction task simultaneously. Given the synthetic scene, the first task aims to maximize agreement between a pair of synthetic images generated by our proposed view sampling module, while the second task aims to predict three physical property maps, i.e., depth, instance contour maps, and surface normal maps. In addition, a feature-level domain adaptation technique with adversarial training is applied to reduce the domain difference between the realistic and the synthetic data. Experiments demonstrate that our proposed method achieves state-of-the-art performance on several visual recognition datasets.
42
Geng, Shijie, Peng Gao, Moitreya Chatterjee, Chiori Hori, Jonathan Le Roux, Yongfeng Zhang, Hongsheng Li, and Anoop Cherian. "Dynamic Graph Representation Learning for Video Dialog via Multi-Modal Shuffled Transformers." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 2 (May 18, 2021): 1415–23. http://dx.doi.org/10.1609/aaai.v35i2.16231.
Повний текст джерелаАнотація:
Given an input video, its associated audio, and a brief caption, the audio-visual scene aware dialog (AVSD) task requires an agent to indulge in a question-answer dialog with a human about the audio-visual content. This task thus poses a challenging multi-modal representation learning and reasoning scenario, advancements into which could influence several human-machine interaction applications. To solve this task, we introduce a semantics-controlled multi-modal shuffled Transformer reasoning framework, consisting of a sequence of Transformer modules, each taking a modality as input and producing representations conditioned on the input question. Our proposed Transformer variant uses a shuffling scheme on their multi-head outputs, demonstrating better regularization. To encode fine-grained visual information, we present a novel dynamic scene graph representation learning pipeline that consists of an intra-frame reasoning layer producing spatio-semantic graph representations for every frame, and an inter-frame aggregation module capturing temporal cues. Our entire pipeline is trained end-to-end. We present experiments on the benchmark AVSD dataset, both on answer generation and selection tasks. Our results demonstrate state-of-the-art performances on all evaluation metrics.
43
Wikantari, Ni Nyoman Yustini, Laila Hayati, Wahidaturrahmi Wahidaturrahmi, and Baidowi Baidowi. "Analysis of mathematical representation ability on pythagoras theorem reviewed from learning style of junior high school students." Jurnal Pijar Mipa 17, no. 6 (November 30, 2022): 775–81. http://dx.doi.org/10.29303/jpm.v17i6.3311.
Повний текст джерелаАнотація:
This study aims to determine the mathematical representation ability of class IX students with visual, auditory, and kinesthetic learning styles on the Pythagoras Theorem material at SMPN 1 (public junior high school) Gunungsari Indonesia in the 2021/2022 academic year. This type of research is descriptive with a quantitative approach. The populations in this study were all students of class IX SMPN 1 Gunungsari. The research sampling technique used purposive sampling, and the sample was class IX-H. The instruments used are learning style questionnaires, mathematical representation ability tests, and interview guidelines. The results of this study indicate that the mathematical representation ability of students with visual learning styles is 35.56% in the low category, or students have yet to be able to meet the overall visual, symbolic, and verbal representation indicators. Furthermore, the mathematical representation ability of students with auditory and kinesthetic learning styles is 49.86% and 41.87% in the medium category. Respectively, students have been able to meet the indicators of visual and symbolic representation quite well but have yet to be able to fulfill the visual and symbolic representation and verbal representation indicators.
44
Cadieu, Charles F., and Bruno A. Olshausen. "Learning Intermediate-Level Representations of Form and Motion from Natural Movies." Neural Computation 24, no. 4 (April 2012): 827–66. http://dx.doi.org/10.1162/neco_a_00247.
Повний текст джерелаАнотація:
We present a model of intermediate-level visual representation that is based on learning invariances from movies of the natural environment. The model is composed of two stages of processing: an early feature representation layer and a second layer in which invariances are explicitly represented. Invariances are learned as the result of factoring apart the temporally stable and dynamic components embedded in the early feature representation. The structure contained in these components is made explicit in the activities of second-layer units that capture invariances in both form and motion. When trained on natural movies, the first layer produces a factorization, or separation, of image content into a temporally persistent part representing local edge structure and a dynamic part representing local motion structure, consistent with known response properties in early visual cortex (area V1). This factorization linearizes statistical dependencies among the first-layer units, making them learnable by the second layer. The second-layer units are split into two populations according to the factorization in the first layer. The form-selective units receive their input from the temporally persistent part (local edge structure) and after training result in a diverse set of higher-order shape features consisting of extended contours, multiscale edges, textures, and texture boundaries. The motion-selective units receive their input from the dynamic part (local motion structure) and after training result in a representation of image translation over different spatial scales and directions, in addition to more complex deformations. These representations provide a rich description of dynamic natural images and testable hypotheses regarding intermediate-level representation in visual cortex.
45
Komala, Elsa, and Asri Maulani Afrida. "Analisis Kemampuan Representasi Matematis Siswa SMK Ditinjau dari Gaya Belajar." Journal of Instructional Mathematics 1, no. 2 (November 16, 2020): 53–59. http://dx.doi.org/10.37640/jim.v1i2.364.
Повний текст джерелаАнотація:
This research aims to describe the mathematical representation abilities of vocational school students in terms of visual, auditory, and kinesthetic learning styles, as well as learning styles that have the best representational abilities in mathematics learning. The research was conducted at SMK Negeri 2 Cilaku Cianjur. The research method used is descriptive research with a quantitative approach. The subjects in this study were all 29 students of class X TKJ 2 with purposive sampling technique. The data used are written tests to reveal mathematical representation abilities, observation and questionnaires to classify students based on learning styles, interviews with students. Data processing used descriptive analysis of the percentage of posttest scores, learning styles by looking at the percentage of observation statements and answers to student questionnaire statements. The results of the data analysis showed that the percentage of achievement of the mathematical representation ability of students with a visual learning style was 71.43% in the sufficient category, students with the auditory learning style 71.25% in the sufficient category, and students with the kinesthetic learning style 73.89% with the sufficient category. The kinesthetic learning style has the best representation ability in mathematics learning with a percentage of 73.89% with a sufficient category.
46
Sabrina, Atika Nur, Isti Hidayah, and Muhammad Kharis. "Mathematical representation ability and confident character assisted by Schoology with the NHT method and thematic approach." Unnes Journal of Mathematics Education 10, no. 2 (August 31, 2021): 91–98. http://dx.doi.org/10.15294/ujme.v10i2.29317.
Повний текст джерелаАнотація:
This study aims to determine the students’ mathematical representation ability and confidence character assisted by Schoology with the Numbered Heads Together (NHT) method and thematic approach and describe the mathematical representation ability based on the character of self-confidence. The subjects of this study were students of class VIIF and VIIH, one of Junior High School in Semarang. The method used is a mixed method. The results of the study show that: (1) The students’ mathematical representations ability and confident character assisted by Schoology with the NHT method and thematic approaches are better than learning without Schoology; (2) the mathematical representation ability based on the character of self-confidence is (a) students with high self-confidence implement indicators visual and symbolic representations; (b) students with medium self-confidence implement indicator symbolic representation; and (c) students with low self-confidence not implementing of indicators visual, symbolic, and verbal representation abilities.
47
Mustapha, Abolaji Samuel. "Dynamics of gender representations in learning materials and gender equality." Multidisciplinary Journal of Gender Studies 1, no. 3 (October 25, 2012): 243–70. http://dx.doi.org/10.4471/generos.2012.12.
Повний текст джерелаАнотація:
The concept of representation has been taken up in many disciplines, largely, in visual arts, music, media studies, feminism, gender studies among others.The particular interest that researchers in gender and education studies have taken in gender representation has yielded many studies that have in turn reported interesting findings that are instrumentals to revision of learning materials and education process/programmes in some countries.This paper explicates on the concept of representation and its dynamics through learning materials in order to stress a need for studies in under-researched sites and to shedlight on the importance of gender representation in text studies with the anticipation that those who do not share this orientation, and whose take on the concern with gender representation in learning materials studies is that it is a nonessential issue in education mightbe able to appreciate both the undertakings and the findings of studies on gender representations in textbooks.Keywords: representation, gender equality, socialisation, learning materials, education
48
Taqwa, Muhammad Reyza Arief, and Handy Faishal Rahim. "Students’ conceptual understanding on vector topic in visual and mathematical representation: a comparative study." Journal of Physics: Conference Series 2309, no. 1 (July 1, 2022): 012060. http://dx.doi.org/10.1088/1742-6596/2309/1/012060.
Повний текст джерелаАнотація:
Abstract This study aims to compare students’ abilities in understanding vector topics in two representations, namely visual and mathematical. This is a comparative research. Data were collected by survey. The research was conducted on 191 first-year undergraduate students of physics and physics education. The research instrument consisted of 14 multiple choice questions (7 questions of mathematical representation and 7 questions of verbal representation). Data analysis was performed by determining descriptive statistics and paired sample t-test. The results showed that the students’ ability to understand vector concepts with mathematical representations was better than those in verbal representations. Students’ mean score in verbal and mathematical representation formats are 33.91 and 59.17. Based on the results of the paired sample t-test obtained t = −12.96 and sig. = 0.00. These results indicate that the students’ ability to understand vector concepts in verbal and mathematical representation formats is significantly different. This study’s results came that students’ understanding of vector concepts still depends on the representation of the questions because their understanding is not coherent. Based on these findings, vector learning should be focused on the meaning of vectors in various representations, and connecting the meanings of vector operations in various representations, not only on mathematics.
49
Winarti, A., A. Almubarak, and P. Saadi. "Visual learning style-based chemistry mental model representation through transformative online learning." Journal of Physics: Conference Series 2104, no. 1 (November 1, 2021): 012023. http://dx.doi.org/10.1088/1742-6596/2104/1/012023.
Повний текст джерелаАнотація:
Abstract Analysis of learning styles is one of the main things teachers need to do before carrying out teaching. The study of learning styles can provide an overview of how a teacher designs a learning concept by students’ learning styles. The learning process will show students’ mental models with their respective learning styles so that these mental models become the primary material for how teachers develop students’ cognitive. This research aimed to describe students’ mental models in terms of students’ visual learning styles. The method used is descriptive with qualitative and quantitative approaches with transformative-based learning concepts. The research results show that chemistry education students for chemistry learning innovation courses only have visual learning styles of 71.43% and audio by 28.55% and do not have kinaesthetic learning styles. This research focuses on visual learning styles to see students’ mental models. The conclusion is that students still need cognitive strengthening, especially the ability to interpret phenomena at the sub microscopic level. With the visual learning style, students are expected to transform their cognitive so that they have mental structures and models relevant in theory and terminology.
50
Wasis, W. Widodo, T. Sunarti, W. Setyarsih, M. N. R. Jauhariyah, and A. Zainuddin. "The Relationship between Multiple Representational Skills and Understanding of Physics Concepts in the Pre-Service Science Teacher." Journal of Physics: Conference Series 2623, no. 1 (November 1, 2023): 012031. http://dx.doi.org/10.1088/1742-6596/2623/1/012031.
Повний текст джерелаАнотація:
Abstract The concept of physics has various representations that must be mastered by the Pre-Service Science Teacher (PSST). It is not uncommon for PSST to have poor multiple representation skills, impacting the delivery of concepts in school. Therefore, this study aims to analyze the profile of multiple representation skills and their relation to understanding the concept of PSST Physics. The representations studied include verbal, visual, symbolic, and mathematical forms. Eleven PSST became respondents in the study. Multiple representation and conception measurements use instruments with a three-tier item format containing content, argumentation, and confidence levels. Student responses are analyzed descriptively, qualitatively, and quantitatively. The findings of this study are: 1) 18% of PSST belong to the concept understanding, 27% experienced misconceptions, and the rest (55%) were classified as not knowing the concept or responding by guessing; 2) the highest representation format mastered by students is the visual representation, and the lowest is the mathematical representation; and 3) the ability to multiple representations and understand physics concepts has a significant and perfect correlation with a Pearson Correlation of 0.847. This research implies that learning for PSST can emphasize multiple representation abilities as it affects their understanding of concepts.