Zobrazeno 1 - 10
of 32
pro vyhledávání: '"Scott Niekum"'
Publikováno v:
Hanna, J P, Niekum, S & Stone, P 2021, ' Importance sampling in reinforcement learning with an estimated behavior policy ', Machine Learning, vol. 110, no. 6, pp. 1267-1317 . https://doi.org/10.1007/s10994-020-05938-9
In reinforcement learning, importance sampling is a widely used method for evaluating an expectation under the distribution of data of one policy when the data has in fact been generated by a different policy. Importance sampling requires computing t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e1fb57fcacfc7d80f887482c30e39994
https://doi.org/10.1007/s10994-020-05938-9
https://doi.org/10.1007/s10994-020-05938-9
Autor:
Aaron Steinfeld, Reid Simmons, Scott Niekum, Pallavi Koppol, Yuchen Cui, Tesca Fitzgerald, Henny Admoni
Publikováno v:
IJCAI
Human-in-the-loop Machine Learning (HIL-ML) is a widely adopted paradigm for instilling human knowledge in autonomous agents. Many design choices influence the efficiency and effectiveness of such interactive learning processes, particularly the inte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d37126003855b79dff46b07b84dbffe3
https://doi.org/10.24963/ijcai.2021/599
https://doi.org/10.24963/ijcai.2021/599
Autor:
Scott Niekum, Daniel S. Brown
Publikováno v:
AAAI
Inverse reinforcement learning (IRL) infers a reward function from demonstrations, allowing for policy improvement and generalization. However, despite much recent interest in IRL, little work has been done to understand the minimum set of demonstrat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fa4de9d559b5f11500cbfeda5abd532e
https://doi.org/10.1609/aaai.v33i01.33017749
https://doi.org/10.1609/aaai.v33i01.33017749
Publikováno v:
ICRA
Robots in human environments will need to interact with a wide variety of articulated objects such as cabinets, drawers, and dishwashers while assisting humans in performing day-to-day tasks. Existing methods either require objects to be textured or
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::888536c38a929c29d6474b9f5dc4a753
https://doi.org/10.1109/icra48506.2021.9561132
https://doi.org/10.1109/icra48506.2021.9561132
Autor:
Scott Niekum
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 26:2402-2403
Much work in learning from demonstration has focused on learning simple tasks from structured demonstrations that have a well-defined beginning and end. As we attempt to scale robot learning to increasingly complex tasks, it becomes intractable to le
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::02d357ca75de8f0f9e306ec768241c4b
https://doi.org/10.1609/aaai.v26i1.8182
https://doi.org/10.1609/aaai.v26i1.8182
Publikováno v:
IROS
Deep reinforcement learning (DRL) is capable of learning high-performing policies on a variety of complex high-dimensional tasks, ranging from video games to robotic manipulation. However, standard DRL methods often suffer from poor sample efficiency
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ee9d71e3bbe3fb18f08a585b1943941e
https://doi.org/10.1109/iros45743.2020.9340891
https://doi.org/10.1109/iros45743.2020.9340891
Autor:
Bo Liu, Ruohan Zhang, Dana H. Ballard, Scott Niekum, Mary Hayhoe, Yifeng Zhu, Akanksha Saran, Sihang Guo
Publikováno v:
IJCAI (U S)
IJCAI
IJCAI
Human gaze reveals a wealth of information about internal cognitive state. Thus, gaze-related research has significantly increased in computer vision, natural language processing, decision learning, and robotics in recent years. We provide a high-lev
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::da4d724493795a8676e83204ca5bcbb6
https://pubmed.ncbi.nlm.nih.gov/32901189
https://pubmed.ncbi.nlm.nih.gov/32901189
Autor:
Scott Niekum, Wonjoon Goo
Publikováno v:
ICRA
Due to burdensome data requirements, learning from demonstration often falls short of its promise to allow users to quickly and naturally program robots. Demonstrations are inherently ambiguous and incomplete, making correct generalization to unseen
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7c0783f560d7a39fd788fcfd7a22afae
https://doi.org/10.1109/icra.2019.8793515
https://doi.org/10.1109/icra.2019.8793515
Publikováno v:
ICRA
Estimating statistical uncertainties allows autonomous agents to communicate their confidence during task execution and is important for applications in safety-critical domains such as autonomous driving. In this work, we present the uncertainty-awar
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c55ab88c0c13dd73911d9d663f3e24a0
https://doi.org/10.1109/icra.2019.8794025
https://doi.org/10.1109/icra.2019.8794025
Publikováno v:
IJCAI
Recent reinforcement learning (RL) approaches have shown strong performance in complex domains such as Atari games, but are often highly sample inefficient. A common approach to reduce interaction time with the environment is to use reward shaping, w
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::26b5a8a48e049ecbd1df018332d66770
http://arxiv.org/abs/1903.02020
http://arxiv.org/abs/1903.02020
