Zobrazeno 1 - 10
of 44
pro vyhledávání: '"Mandal, Debmalya"'
We consider the problem of recovering the ground truth ordering (ranking, top-$k$, or others) over a large number of alternatives. The wisdom of crowd is a heuristic approach based on Condorcet's Jury theorem to address this problem through collectiv
Externí odkaz:
http://arxiv.org/abs/2406.00870
High-dimensional linear bandits with low-dimensional structure have received considerable attention in recent studies due to their practical significance. The most common structure in the literature is sparsity. However, it may not be available in pr
Externí odkaz:
http://arxiv.org/abs/2405.13899
We study data corruption robustness in offline two-player zero-sum Markov games. Given a dataset of realized trajectories of two players, an adversary is allowed to modify an $\epsilon$-fraction of it. The learner's goal is to identify an approximate
Externí odkaz:
http://arxiv.org/abs/2403.07933
Autor:
Nika, Andi, Mandal, Debmalya, Kamalaruban, Parameswaran, Tzannetos, Georgios, Radanović, Goran, Singla, Adish
In this paper, we take a step towards a deeper understanding of learning from human preferences by systematically comparing the paradigm of reinforcement learning from human feedback (RLHF) with the recently proposed paradigm of direct preference opt
Externí odkaz:
http://arxiv.org/abs/2403.01857
When Reinforcement Learning (RL) agents are deployed in practice, they might impact their environment and change its dynamics. We propose a new framework to model this phenomenon, where the current environment depends on the deployed policy as well a
Externí odkaz:
http://arxiv.org/abs/2402.09838
Reward allocation, also known as the credit assignment problem, has been an important topic in economics, engineering, and machine learning. An important concept in reward allocation is the core, which is the set of stable allocations where no agent
Externí odkaz:
http://arxiv.org/abs/2402.07067
We study data corruption robustness for reinforcement learning with human feedback (RLHF) in an offline setting. Given an offline dataset of pairs of trajectories along with feedback about human preferences, an $\varepsilon$-fraction of the pairs is
Externí odkaz:
http://arxiv.org/abs/2402.06734
Establishing causal relationships between actions and outcomes is fundamental for accountable multi-agent decision-making. However, interpreting and quantifying agents' contributions to such relationships pose significant challenges. These challenges
Externí odkaz:
http://arxiv.org/abs/2310.11334
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence 37(10) (2023) 11980-11988
Canonical models of Markov decision processes (MDPs) usually consider geometric discounting based on a constant discount factor. While this standard modeling approach has led to many elegant results, some recent studies indicate the necessity of mode
Externí odkaz:
http://arxiv.org/abs/2307.10491
We study a sequential decision making problem between a principal and an agent with incomplete information on both sides. In this model, the principal and the agent interact in a stochastic environment, and each is privy to observations about the sta
Externí odkaz:
http://arxiv.org/abs/2306.03832