5,643 Hits in 8.8 sec

Learning to Expand: Reinforced Pseudo-relevance Feedback Selection for Information-seeking Conversations [article]

Haojie Pan, Cen Chen, Minghui Qiu, Liu Yang, Feng Ji, Jun Huang, Haiqing Chen
2020 arXiv   pre-print
Intelligent personal assistant systems for information-seeking conversations are increasingly popular in real-world applications, especially for e-commerce companies.  ...  With the development of research in such conversation systems, the pseudo-relevance feedback (PRF) has demonstrated its effectiveness in incorporating relevance signals from external documents.  ...  CONCLUSION In this work, we propose a principled way to automatically select useful pseudo-relevance feedback terms to help information-seeking conversations.  ... 
arXiv:2011.12771v1 fatcat:frpxvpshr5bb3jksiqrkv7lzku

Reinforcement Learning-based Dialogue Guided Event Extraction to Exploit Argument Relations [article]

Qian Li, Hao Peng, Jianxin Li, Jia Wu, Yuanxing Ning, Lihong Wang, Philip S. Yu, Zheng Wang
2021 arXiv   pre-print
To model the argument relation, we employ reinforcement learning and incremental learning to extract multiple arguments via a multi-turned, iterative process.  ...  This two-way feedback process allows us to exploit the argument relations to effectively settle argument roles, leading to better sentence understanding and event extraction.  ...  to extract event triggers and argument roles. for the argument role selected by the reinforcement learning- JMEE [6], a jointly event extraction framework, introduces based dialogue management  ... 
arXiv:2106.12384v2 fatcat:blyylym77vdupbrolil2dtmrna

Applications of the Free Energy Principle to Machine Learning and Neuroscience [article]

Beren Millidge
2021 arXiv   pre-print
reinforcement learning methods.  ...  We go on to propose novel and simpler algorithms which allow for backprop to be implemented in purely local, biologically plausible computations.  ...  Conversely, the fact that evidence objectives seek to minimize the entropy of the predicted distribution means that they implicitly seek to minimize the amount of mutual information between observation  ... 
arXiv:2107.00140v1 fatcat:c6phd65xwfc2rcyq7pnth5a3pq

An Introduction to Lifelong Supervised Learning [article]

Shagun Sodhani, Mojtaba Faramarzi, Sanket Vaibhav Mehta, Pranshu Malviya, Mohamed Abdelsalam, Janarthanan Janarthanan, Sarath Chandar
2022 arXiv   pre-print
desiderata for an ideal lifelong learning system (Section 2.6), discuss how lifelong learning is related to other learning paradigms (Section 2.7), describe common metrics used to evaluate lifelong learning  ...  This chapter is more useful for readers who are new to lifelong learning and want to get introduced to the field without focusing on specific approaches or benchmarks.  ...  for sampling transitions in incremental reinforcement learning.  ... 
arXiv:2207.04354v2 fatcat:yy7sxiwakrcgdleaylaxxri22m

Tutorial on amortized optimization for learning to optimize over continuous domains [article]

Brandon Amos
2022 arXiv   pre-print
This framing enables us easily see, for example, that the amortized inference in variational autoencoders is conceptually identical to value gradients in control and reinforcement learning as they both  ...  We then view existing applications through these foundations to draw connections between them, including for manifold optimization, variational inference, sparse coding, meta-learning, control, reinforcement  ...  , and Atlas Wang for insightful discussions and feedback on this tutorial.  ... 
arXiv:2202.00665v2 fatcat:tvgy2hp2i5f23cwbhuoex3gw2m

Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems

Liu Yang, Minghui Qiu, Chen Qu, Jiafeng Guo, Yongfeng Zhang, W. Bruce Croft, Jun Huang, Haiqing Chen
2018 The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval - SIGIR '18  
We incorporate external knowledge into deep neural models with pseudo-relevance feedback and QA correspondence knowledge distillation.  ...  In this paper, we propose a learning framework on the top of deep neural matching networks that leverages external knowledge for response ranking in information-seeking conversation systems.  ...  For future work, we plan to model user intent in information-seeking conversations and learn meaningful patterns from user intent dynamics to help response selection.  ... 
doi:10.1145/3209978.3210011 dblp:conf/sigir/YangQQGZCHC18 fatcat:xt6767facjezpd6umj3ojrrgh4

From self-tuning regulators to reinforcement learning and back again [article]

Nikolai Matni, Alexandre Proutiere, Anders Rantzer, Stephen Tu
2019 arXiv   pre-print
The goal of this tutorial paper is to provide a starting point for control theorists wishing to work on learning related problems, by covering recent advances bridging learning and control theory, and  ...  Machine and reinforcement learning (RL) are increasingly being applied to plan and control the behavior of autonomous systems interacting with the physical world.  ...  ADP) or, as it has come to be known, reinforcement learning.  ... 
arXiv:1906.11392v2 fatcat:bsmfp526c5b4rpnsgpkdb2fksu

A systematic review of research on the use of technology-supported cooperative learning to enhance self-directed learning [chapter]

Elsa Mentz, North-West University, Roxanne Bailey, North-West University
2019 NWU Self-Directed Learning Series  
Acknowledgements The author is grateful for the funding received from the NRF and the Fuchs Foundation.  ...  This research is part of the NRF-funded project Multimodal multiliteracies in support of self-directed learning.  ...  learning control, and to seek more frequently a greater variety of feedback types.  ... 
doi:10.4102/aosis.2019.bk134.07 fatcat:axfyl57vk5a3xgfryjtpqzej54

Machine-Learning Techniques [chapter]

Rob Sullivan
2011 Introduction to Data Mining for the Life Sciences  
, for motor control (Inverse Kinematics), for learning sensory-motor patterns, and for simulating biologically plausible computational processes.  ...  The tutorial and exercises that will accompany the class with revolve around a number of realworld applications of Machine Learning, namely, for visual processing (face recognition, object recognition)  ...  the algorithm learns a model that best represents a set of inputs without any feedback (no desired output, no external reinforcement)  Learning to learnwhere the algorithm learns its own inductive bias  ... 
doi:10.1007/978-1-59745-290-8_8 fatcat:svpyv2kctzfgthsu3t5rsc575u

Cost-to-Go Function Approximation [chapter]

2017 Encyclopedia of Machine Learning and Data Mining  
Adaptations to the multi-class case are typically performed via class binarization, learning different rule sets for binary problems.  ...  If there are any positive examples in the training set, it calls the subroutine FINDBESTRULE for learning a single rule that Nilsson NJ (1995) Reacting, planning and learning in an autonomous agent.  ...  Case Retrieval Nets are designed to speed up retrieval by applying spreading activation to select relevant cases.  ... 
doi:10.1007/978-1-4899-7687-1_100093 fatcat:vse7ncdqs5atlosjhz7fhlj3im

A Relational Approach to Tool-Use Learning in Robots [chapter]

Solly Brown, Claude Sammut
2013 Lecture Notes in Computer Science  
Acknowledgements I'd firstly like to thank my supervisor, Claude Sammut, for his advice and guidance during the course of this PhD.  ...  Claude encouraged me to explore a new and interesting area in this research, and has always been supportive of my work. Thank-you Claude!  ...  Salganicoff et al. (1996) uses a version of ID3 decision trees to learn to select useful approach directions for object grasping from visual information.  ... 
doi:10.1007/978-3-642-38812-5_1 fatcat:2mmyfk5rireelksnfp2vljktmu

Teaching Law to Accounting and Business Students: A Cumulative Dual Model

F. Ewang, Charles Sturt University
2008 Journal of University Teaching and Learning Practice  
The purpose is to compare and evaluate efficacy of a traditional, lecture-based learning (LBL)i with a combination of LBL and problem-based learning (PBL)ii in improving performance and outcomes for students  ...  This article presents a reflection and comparison of two of my teaching pedagogical approaches for the Business Organisations Law curriculum to undergraduate non-law students at Charles Sturt University  ...  This approach was used because feedback from students was a useful source of information for the structure and revision of the proceeding session.  ... 
doi:10.53761/ fatcat:ljbct5xuhreuja4n4zvhjsehti

Machine learning approaches to power-system security assessment

L. Wehenkel
1997 IEEE Expert  
The two last chapters of this part describe an in depth investigation of the data base generation techniques appropriate for different types of physical problems.  ...  Henri Louis Bergson (1859-1941) A book should have either intelligibility or correctness; to combine the two is impossible.  ...  To exploit them properly, statistical learning techniques are used to extract the relevant information.  ... 
doi:10.1109/64.621229 fatcat:33wtxrctsbcxvjhpunwwm4vmh4

A comprehensive dimensional reduction framework to learn single-cell phenotypic topology uncovers T cell diversity [article]

Davi Sidarta-Oliveira, Licio A Velloso
2022 bioRxiv   pre-print
Here, we dissected DR steps to build TopOMetry, a machine learning framework that learns latent data topology to perform DR in a modular fashion and show that current analysis practices are biased due  ...  We hope this powerful tool will accelerate biomedical discovery and inspire new methods to learn and explore phenotypic topology.  ...  We thank Ebru Erbay and Helder Nakaya for their generous feedback during the development of TopOMetry.  ... 
doi:10.1101/2022.03.14.484134 fatcat:ctj66yfd7fcadc3xbhhmhlwkty

Dynamic Search – Optimizing the Game of Information Seeking [article]

Zhiwen Tang, Grace Hui Yang
2021 arXiv   pre-print
The article reviews approaches to modeling dynamics during information seeking, with an emphasis on Reinforcement Learning (RL)-enabled methods.  ...  Details are given for how different approaches are used to model interactions among the human user, the search system, and the environment.  ...  ACKNOWLEDGMENTS The authors would like to thank Ian Soboroff, Jiyun Luo, Shiqi Liu, Angela Yang, and Xuchu Dong for their past efforts during our long-term collaboration on dynamic search.  ... 
arXiv:1909.12425v2 fatcat:mdby4xq4jrg4pm6jpxslsv2sri
« Previous Showing results 1 — 15 out of 5,643 results