154,679 Hits in 3.3 sec

Toward Causal Representation Learning

Bernhard Scholkopf, Francesco Locatello, Stefan Bauer, Nan Rosemary Ke, Nal Kalchbrenner, Anirudh Goyal, Yoshua Bengio
2021 Proceedings of the IEEE  
A central problem for AI and causality is, thus, causal representation learning, that is, the discovery of highlevel causal variables from low-level observations.  ...  | The two fields of machine learning and graphical causality arose and are developed separately.  ...  Causal representation learning should move beyond the representation of statistical dependence structures toward models that support intervention, planning, and reasoning, realizing Konrad Lorenz' notion  ... 
doi:10.1109/jproc.2021.3058954 fatcat:jqg6jm2f35aynlszy6w5nsyem4

Towards Causal Representation Learning [article]

Bernhard Schölkopf, Francesco Locatello, Stefan Bauer, Nan Rosemary Ke, Nal Kalchbrenner, Anirudh Goyal, Yoshua Bengio
2021 arXiv   pre-print
A central problem for AI and causality is, thus, causal representation learning, the discovery of high-level causal variables from low-level observations.  ...  The two fields of machine learning and graphical causality arose and developed separately.  ...  ACKNOWLEDGMENTS Many thanks to the past and present members of the Tübingen causality team, without whose work and insights this article would not exist, in particular to Dominik Janzing, Chaochao Lu and  ... 
arXiv:2102.11107v1 fatcat:n25xwac72nfulgl3gvvs4kerca

Towards Learning Causal Representations from Multi-Instance Bags [article]

Weijia Zhang, Xuanhui Zhang, Hanwen Deng, Min-Ling Zhang
2022 arXiv   pre-print
This work studies MIL from a new perspective by considering bags as important auxiliary information that can be utilized to identify invariant causal representations from bag-level weak supervision.  ...  Multi-instance learning (MIL) deals with objects represented as bags of instances and can predict instance labels from bag-level supervision.  ...  IRM with Game Theory (IRM GAME) [1] , Invariant Causal Representation Learning (iCaRL) [18] .  ... 
arXiv:2202.12570v2 fatcat:gohroqgazvbtfae2j4jtmqtdwe

Towards Learning Causal Representations of Technical Word Embeddings for Smart Troubleshooting

Alexandre Trilla, Nenad Mijatovic, Xavier Vilasis-Cardona
2022 International Journal of Prognostics and Health Management  
This work explores how the causality inference paradigm may be applied to troubleshoot the root causes of failures through language processing and Deep Learning.  ...  of the underlying causal semantics.  ...  causal effects and counterfactual outcomes with DL is to learn representations for features, i.e., to let the DL system automatically discover the most effective way to represent the data directly instead  ... 
doi:10.36001/ijphm.2022.v13i2.3127 fatcat:r2s6hcvaqrb3znbwy3exv4mbuu

Giving the Giggles: Prediction, Intervention, and Young Children's Representation of Psychological Events

Paul Muentener, Daniel Friel, Laura Schulz, Tiziana Zalla
2012 PLoS ONE  
We discuss these findings as they bear on the development of causal concepts.  ...  Moreover, causal representations crosscut conceptual boundaries.  ...  The conditions under which toddlers fail to form causal representations may violate these early expectations of contact causality.  ... 
doi:10.1371/journal.pone.0042495 pmid:22916130 pmcid:PMC3423398 fatcat:s2qbmomj7vedpn52pcj643e4ty

Disentangling User Interest and Conformity for Recommendation with Causal Embedding [article]

Yu Zheng, Chen Gao, Xiang Li, Xiangnan He, Depeng Jin, Yong Li
2021 arXiv   pre-print
In this paper, we present DICE, a general framework that learns representations where interest and conformity are structurally disentangled, and various backbone recommendation models could be smoothly  ...  However, observational interaction data could result from users' conformity towards popular items, which entangles users' real interest.  ...  Disentangled Representation Learning In this section, we elaborate our designs on disentangling the two causal embeddings for interest and conformity.  ... 
arXiv:2006.11011v2 fatcat:q6m5efzoz5fdlmpfx4fikx43em

Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective [article]

Yuejiang Liu, Riccardo Cadei, Jonas Schweizer, Sherwin Bahmani, Alexandre Alahi
2022 arXiv   pre-print
In this work, we propose to address these challenges from a causal representation perspective.  ...  approximate a sparse causal graph; (iii) we introduce a style contrastive loss that not only enforces the structure of style representations but also serves as a self-supervisory signal for test-time  ...  We hope our findings will pave the way for a tight integration of causal modeling and representation learning in the motion context, a largely under-explored yet highly promising direction towards reliable  ... 
arXiv:2111.14820v4 fatcat:u34zervh4nakrbzc7uoqmtwqyq

Page 2372 of Psychological Abstracts Vol. 81, Issue 5 [page]

1994 Psychological Abstracts  
Toward a theory of learning and repre- senting causal inferences in neural networks. [In: (PA Vol 81: 19806) Neural networks for knowledge representation and infer- ence.  ...  be considered a basic form of causal relation learning review some methods which have been employed to incor- porate temporal representation in neural networks © [intro- duce] a novel algorithm for producing  ... 

Model-Based Methods for Assessment, Learning, and Instruction: Innovative Educational Technology at Florida State University [chapter]

Valerie J. Shute, Allan C. Jeong, J. Michael Spector, Norbert M. Seel, Tristan E. Johnson
2009 Educational Media and Technology Yearbook  
We call our approach model-based because it integrates representations of mental models and internal cognitive processes with tools that are used to (a) assess progress of learning, and (b) provide the  ...  Our methods and tools represent an approach to learning and instruction that is now embedded in many of the graduate courses at Florida State University and also at the University of Freiburg.  ...  The reason for using causal representations as the basis for analysis is that such representations reflect internal relationships among factors and components (i.e., problem dynamics), and causal representations  ... 
doi:10.1007/978-0-387-09675-9_5 fatcat:aepj27afwbd7nczlrlh3laftlq

CausalKG: Causal Knowledge Graph Explainability using interventional and counterfactual reasoning [article]

Utkarshani Jaimini, Amit Sheth
2022 arXiv   pre-print
The proposed Causal Knowledge Graph (CausalKG) framework, leverages recent progress of causality and KG towards explainability.  ...  The causality representation requires a higher representation framework to define the context, the causal information, and the causal effects.  ...  This research is support in part by National Science Foundation (NSF) Award # 2133842 "EAGER: Advancing Neurosymbolic AI with Deep Knowledge-infused Learning," and Award #2119654, "RII Track 2 FEC: Enabling  ... 
arXiv:2201.03647v1 fatcat:kqeoqjsdz5dgjil2aqjj65ehzu

Page 222 of Journal of Science Education and Technology Vol. 4, Issue 3 [page]

1995 Journal of Science Education and Technology  
Toward a Causal Model The single most important development re- vealed across the students’ performances on the posttest items was a major movement toward un- derstanding the interrelation of the components  ...  Prior to the classroom learn- ing, half the students had an idea of reflectance as a necessary condition for visibility, but the causal link between light source and rays, the ob- jects in the room as  ... 

Towards Causal Knowledge Graphs - Position Paper

Eva Blomqvist, Marjan Alirezaie, Marina Santini
2020 European Conference on Artificial Intelligence  
However, most approaches purely based on Machine Learning (ML) do not explicitly represent and reason with causal relations, and may therefore mistake correlation for causation.  ...  to other types of relations, apart from causality.  ...  to involve causal models within a learning process.  ... 
dblp:conf/ecai/BlomqvistAS20 fatcat:jc5jyx3kujck5mj4lkmxbbb75u

Invariant Grounding for Video Question Answering [article]

Yicong Li, Xiang Wang, Junbin Xiao, Wei Ji, Tat-Seng Chua
2022 arXiv   pre-print
Towards this end, we propose a new learning framework, Invariant Grounding for VideoQA (IGV), to ground the question-critical scene, whose causal relations with answers are invariant across different interventions  ...  In this work, we first take a causal look at VideoQA and argue that invariant grounding is the key to ruling out the spurious correlations.  ...  Towards this end, we propose a new learning framework, Invariant Grounding for VideoQA (IGV).  ... 
arXiv:2206.02349v1 fatcat:ik47bkqqvrbqfgqzie55h4xfce

Page 483 of Journal of Science Education and Technology Vol. 12, Issue 4 [page]

2003 Journal of Science Education and Technology  
Mecha- nistic cues hinting at how to represent components and causal relations between components would lead to more integrated causal models made up of subcomponent models in mental representation.  ...  Journal of Science Education and Technology, Vol. 12, No. 4, December 2003 (© 2003) Mental Models and Computer-Based Scientific Inquiry Learning: Effects of Mechanistic Cues on Adolescent Representation  ... 

The case of CAUSE: neurobiological mechanisms for grounding an abstract concept

Friedemann Pulvermüller
2018 Philosophical Transactions of the Royal Society of London. Biological Sciences  
One contribution of 23 to a theme issue 'Varieties of abstract concepts: development, use and representation in the brain'.  ...  In this model, Figure 3 . 3 Different theories of causal learning and representation put emphasis on correlations between perceptions ( perception model), or on the learning of motor acts and their typical  ...  Only the mirror-Piaget model explains causal learning by observation along with the 'action advantage' in causal learning (see text for discussion). Phil. Trans. R.  ... 
doi:10.1098/rstb.2017.0129 pmid:29914997 pmcid:PMC6015827 fatcat:s3y5dc7n65fblmwwvmintzubca
« Previous Showing results 1 — 15 out of 154,679 results