Filters








14,263 Hits in 4.6 sec

Systematic Evaluation of Causal Discovery in Visual Model Based Reinforcement Learning [article]

Nan Rosemary Ke, Aniket Didolkar, Sarthak Mittal, Anirudh Goyal, Guillaume Lajoie, Stefan Bauer, Danilo Rezende, Yoshua Bengio, Michael Mozer, Christopher Pal
2021 arXiv   pre-print
We evaluate various representation learning algorithms from the literature and find that explicitly incorporating structure and modularity in models can help causal induction in model-based reinforcement  ...  A central goal for AI and causality is thus the joint discovery of abstract representations and causal structure.  ...  We would also like to thank the developers of Pytorch for developments of great frameworks. We would like to thank Dmitriy Serdyuk for useful feedback and discussions.  ... 
arXiv:2107.00848v1 fatcat:gvxnknjlijaaljkujkcpkefc3q

EST: Evaluating Scientific Thinking in Artificial Agents [article]

Manjie Xu, Guangyuan Jiang, Chi Zhang, Song-Chun Zhu, Yixin Zhu
2022 arXiv   pre-print
By evaluating Reinforcement Learning (RL) agents on both a symbolic and visual version of this task, we notice clear failure of today's learning methods in reaching a level of intelligence comparable to  ...  Motivated by the stream of research on causal discovery, we build our interactive EST environment based on Blicket detection.  ...  Inspired by Blicket detection and the problem of causal induction [18, 19] , the ACRE dataset [50] was presented as a way to systematically evaluate current vision systems' capability in causal induction  ... 
arXiv:2206.09203v1 fatcat:nzjdle2lxvavlp6oa4nzmcsw7a

Data science and AI in FinTech: An overview [article]

Longbing Cao, Qiang Yang, Philip S. Yu
2021 arXiv   pre-print
of smart FinTech futures to the DSAI communities.  ...  blockchain, and the DSAI techniques including complex system methods, quantitative methods, intelligent interactions, recognition and responses, data analytics, deep learning, federated learning, privacy-preserving  ...  -Causality analysis: such as linear and nonlinear Granger causality; causally anomalous multivariate time series; causal tree-based causal inference with instrumental variables; etc.  ... 
arXiv:2007.12681v2 fatcat:jntzuwaktjg2hmmjypi5lvyht4

Neuropathic Pain Diagnosis Simulator for Causal Discovery Algorithm Evaluation [article]

Ruibo Tu, Kun Zhang, Bo Christer Bertilson, Hedvig Kjellström, Cheng Zhang
2019 arXiv   pre-print
In this work, we handle the problem of evaluating causal discovery algorithms by building a flexible simulator in the medical setting.  ...  Our simulator provides a natural tool for evaluating various types of causal discovery algorithms, including those to deal with practical issues in causal discovery, such as unknown confounders, selection  ...  In addition, the authors thank Akshaya Thippur Sridatta and Tino Weinkauf for the help of the audio dubbing of the 3-minute introduction video and the visualization of the causal graph.  ... 
arXiv:1906.01732v3 fatcat:ouyf3xaqp5azjfbnqxzlvmcfqi

Towards a Solution to Bongard Problems: A Causal Approach [article]

Salahedine Youssef and Matej Zečević and Devendra Singh Dhami and Kristian Kersting
2022 arXiv   pre-print
We present a systematic analysis using modern techniques from the intersection of causality and AI/ML in a humble effort of reviving research around BPs.  ...  learning techniques for solving the BPs subject to the causal assumptions.  ...  Acknowledgments This work was supported by the ICT-48 Network of AI Research Excellence Center "TAILOR" (EU Horizon 2020, GA No 952215), the Nexplore Collaboration Lab "AI in Construction" (AICO) and by  ... 
arXiv:2206.07196v1 fatcat:a3wjtp66tvg4noyuuip2y6bwei

Modelling Subject Domain Causality for Learning Content Renewal

Saulius GUDAS, Jurij TEKUTOV, Rimantas BUTLERIS, Vitalijus DENISOVAS
2019 Informatica  
Two levels of the domain causal modelling are obtained. The first level is the discovery of the causality of the domain using the Management Transaction (MT) framework.  ...  This method was used in the field of education, and a case study of learning content renewal is provided. The domain here is a real world area - a learning content is about.  ...  The normalized learning content in Table 3 is evaluated on the level of courses.  ... 
doi:10.15388/informatica.2019.214 fatcat:2dmwrkcjnbg73evw3er6mvtpw4

Automated Learning and Discovery State-of-the-Art and Research Topics in a Rapidly Growing Field

Sebastian Thrun, Christos Faloutsos, Tom M. Mitchell, Larry A. Wasserman
1999 The AI Magazine  
Visual Methods for the Study of Massive Data Sets Organized by Bill Eddy and Steve Eick 6. Learning from Text and the Web Organized by Yiming Yang,  ...  Machine Learning and Reinforcement Learning for Manufacturing Organized by Sridhar Mahadevan and Andrew Moore 4.  ...  Acknowledgments CONALD was held to celebrate CMU's new Center for Automated Learning and Discovery, a cross-disciplinary center created in 1997, whose sponsorship is gratefully acknowledged.  ... 
doi:10.1609/aimag.v20i3.1468 dblp:journals/aim/ThrunFMW99 fatcat:jncvyntf7zetjdnyodz2xmwuyi

Towards Understanding How Machines Can Learn Causal Overhypotheses [article]

Eliza Kosoy, David M. Chan, Adrian Liu, Jasmine Collins, Bryanna Kaufmann, Sandy Han Huang, Jessica B. Hamrick, John Canny, Nan Rosemary Ke, Alison Gopnik
2022 arXiv   pre-print
One of the key challenges for current machine learning algorithms is modeling and understanding causal overhypotheses: transferable abstract hypotheses about sets of causal relationships.  ...  In this work, we present a new benchmark -- a flexible environment which allows for the evaluation of existing techniques under variable causal overhypotheses -- and demonstrate that many existing state-of-the-art  ...  We would like to thank the following children's museums for providing us with a space in which to run the experiments: Bay Area Discovery Museum, Children's Creativity Museum and The Lawrence Hall of Science  ... 
arXiv:2206.08353v1 fatcat:ztllislnkffdnlqf2qrcg5kkwu

D'ya Like DAGs? A Survey on Structure Learning and Causal Discovery

Matthew J. Vowels, Necati Cihan Camgoz, Richard Bowden
2022 ACM Computing Surveys  
Causal reasoning is a crucial part of science and human intelligence. In order to discover causal relationships from data, we need structure discovery methods.  ...  We provide a review of background theory and a survey of methods for structure discovery.  ...  a working deinition of causality and its popular systematization in Structural Causal Models (SCMs).  ... 
doi:10.1145/3527154 fatcat:sroohzvx5reajkia5ythaiyyjm

Explainability in reinforcement learning: perspective and position [article]

Agneza Krajna, Mario Brcic, Tomislav Lipic, Juraj Doncevic
2022 arXiv   pre-print
Reinforcement learning (RL) models increase the space of solvable problems with respect to other machine learning paradigms.  ...  Some of the most interesting applications are in situations with non-differentiable expected reward function, operating in unknown or underdefined environment, as well as for algorithmic discovery that  ...  [55] present a model that generates explanations of behavior based on counterfactual analysis of the structural causal model that is learned during reinforcement learning.  ... 
arXiv:2203.11547v1 fatcat:7zfi7f3i6bdgbhcjg7izvpnieq

A Survey of Deep Reinforcement Learning in Recommender Systems: A Systematic Review and Future Directions [article]

Xiaocong Chen, Lina Yao, Julian McAuley, Guanglin Zhou, Xianzhi Wang
2021 arXiv   pre-print
of the recent trends of deep reinforcement learning in recommender systems.  ...  In light of the emergence of deep reinforcement learning (DRL) in recommender systems research and several fruitful results in recent years, this survey aims to provide a timely and comprehensive overview  ...  [133] propose to use reinforcement learning to search DAG for causal discovery.  ... 
arXiv:2109.03540v2 fatcat:5gwrbfcj3rc7jfkd54eseck5ga

D'ya like DAGs? A Survey on Structure Learning and Causal Discovery [article]

Matthew J. Vowels, Necati Cihan Camgoz, Richard Bowden
2021 arXiv   pre-print
Causal reasoning is a crucial part of science and human intelligence. In order to discover causal relationships from data, we need structure discovery methods.  ...  We provide a review of background theory and a survey of methods for structure discovery.  ...  Causal Discovery with Reinforcement Learning (RL-BIC, 2020) The authors of RL-BIC [272] take a reinforcement learning approach to causal discovery.  ... 
arXiv:2103.02582v2 fatcat:x45blijl5ze5xjyuqh6vlc26oq

Machine learning application in the life time of materials [article]

Xiaojiao Yu
2017 arXiv   pre-print
With the accumulation of data from both experimental and computational results, data based machine learning becomes an emerging field in materials discovery, design and property prediction.  ...  This manuscript reviews the history of materials science as a disciplinary the most common machine learning method used in materials science, and specifically how they are used in materials discovery,  ...  Reinforcement learning is not used in materials science field; hence it is not introduced in detail in this manuscript.  ... 
arXiv:1707.04826v1 fatcat:6wgtrxk2fvgavmbonraip7b4ca

The Big Three: A Methodology to Increase Data Science ROI by Answering the Questions Companies Care About [article]

Daniel K. Griffin
2020 arXiv   pre-print
Companies may be achieving only a third of the value they could be getting from data science in industry applications.  ...  The applications of data science seem to be nearly endless in today's modern landscape, with each company jockeying for position in the new data and insights economy.  ...  Causal discovery algorithms fall loosely into the categories of constraint-based methods and score-based methods.  ... 
arXiv:2002.07069v1 fatcat:if3das3tpbeuba33gmsnmvulrm

Learning Causal Overhypotheses through Exploration in Children and Computational Models [article]

Eliza Kosoy, Adrian Liu, Jasmine Collins, David M Chan, Jessica B Hamrick, Nan Rosemary Ke, Sandy H Huang, Bryanna Kaufmann, John Canny, Alison Gopnik
2022 arXiv   pre-print
Despite recent progress in reinforcement learning (RL), RL algorithms for exploration still remain an active area of research.  ...  Existing methods often focus on state-based metrics, which do not consider the underlying causal structures of the environment, and while recent research has begun to explore RL environments for causal  ...  Causal learning in reinforcement learning There are several standard reinforcement learning benchmarks and environments for causal discovery, including Causal World (Ahmed et al., 2020 ), Causal City  ... 
arXiv:2202.10430v1 fatcat:wvikndmku5bh3e5yxjj5dcecqq
« Previous Showing results 1 — 15 out of 14,263 results