Zobrazeno 1 - 10
of 33
pro vyhledávání: '"Amrit Singh"'
Actor-critic algorithms have shown remarkable success in solving state-of-the-art decision-making problems. However, despite their empirical effectiveness, their theoretical underpinnings remain relatively unexplored, especially with neural network p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0a1463d69b31855af21a4c34ffad7400
http://arxiv.org/abs/2306.10486
http://arxiv.org/abs/2306.10486
Autor:
Patel, Bhrij, Weerakoon, Kasun, Suttle, Wesley A., Koppel, Alec, Sadler, Brian M., Bedi, Amrit Singh, Manocha, Dinesh
Reinforcement learning methods, while effective for learning robotic navigation strategies, are known to be highly sample inefficient. This sample inefficiency comes in part from not suitably balancing the explore-exploit dilemma, especially in the p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fcb9edbf62e713ce817fbe3f95679277
http://arxiv.org/abs/2306.06192
http://arxiv.org/abs/2306.06192
Navigating safely and efficiently in dense and heterogeneous traffic scenarios is challenging for autonomous vehicles (AVs) due to their inability to infer the behaviors or intentions of nearby drivers. In this work, we propose a distributed multi-ag
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b81c1b5f7e7dbe75df0cfed2987e7275
http://arxiv.org/abs/2306.06236
http://arxiv.org/abs/2306.06236
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 36:3682-3689
Reinforcement learning is widely used in applications where one needs to perform sequential decisions while interacting with the environment. The problem becomes more challenging when the decision requirement includes satisfying some safety constrain
Autor:
Chakraborty, Souradip, Bedi, Amrit Singh, Zhu, Sicheng, An, Bang, Manocha, Dinesh, Huang, Furong
Our work focuses on the challenge of detecting outputs generated by Large Language Models (LLMs) to distinguish them from those generated by humans. This ability is of the utmost importance in numerous applications. However, the possibility of such d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::40de4074eac0ac9542ae5313e9f5c937
http://arxiv.org/abs/2304.04736
http://arxiv.org/abs/2304.04736
Autor:
Chakraborty, Souradip, Weerakoon, Kasun, Poddar, Prithvi, Tokekar, Pratap, Bedi, Amrit Singh, Manocha, Dinesh
Reinforcement learning-based policies for continuous control robotic navigation tasks often fail to adapt to changes in the environment during real-time deployment, which may result in catastrophic failures. To address this limitation, we propose a n
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::391557d1e246099533ac929a99f43d9f
http://arxiv.org/abs/2303.07622
http://arxiv.org/abs/2303.07622
Autor:
Suttle, Wesley A., Bedi, Amrit Singh, Patel, Bhrij, Sadler, Brian M., Koppel, Alec, Manocha, Dinesh
Many existing reinforcement learning (RL) methods employ stochastic gradient iteration on the back end, whose stability hinges upon a hypothesis that the data-generating process mixes exponentially fast with a rate parameter that appears in the step-
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::643af96f2803e932e339efb10699773c
Autor:
Chakraborty, Souradip, Bedi, Amrit Singh, Koppel, Alec, Wang, Mengdi, Huang, Furong, Manocha, Dinesh
Directed Exploration is a crucial challenge in reinforcement learning (RL), especially when rewards are sparse. Information-directed sampling (IDS), which optimizes the information ratio, seeks to do so by augmenting regret with information gain. How
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5ea629c25dfb2b65e4b3bfa9610a9b16
Publikováno v:
2022 IEEE 61st Conference on Decision and Control (CDC).
In tabular multi-agent reinforcement learning with average-cost criterion, a team of agents sequentially interacts with the environment and observes local incentives. We focus on the case that the global reward is a sum of local rewards, the joint po
The decentralized Federated Learning (FL) setting avoids the role of a potentially unreliable or untrustworthy central host by utilizing groups of clients to collaboratively train a model via localized training and model/gradient sharing. Most existi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2cd4ca94e7a723a4704351605fba1f03
http://arxiv.org/abs/2210.14026
http://arxiv.org/abs/2210.14026