English
Related papers

Related papers: Robust Q-Learning under Corrupted Rewards

200 papers

We present an approach called Q-probing to adapt a pre-trained language model to maximize a task-specific reward function. At a high level, Q-probing sits between heavier approaches such as finetuning and lighter approaches such as few shot…

Machine Learning · Computer Science 2024-06-04 Kenneth Li , Samy Jelassi , Hugh Zhang , Sham Kakade , Martin Wattenberg , David Brandfonbrener

We study the transfer of rewards learned using inverse reinforcement learning from expert demonstrations in one environment to reinforcement learning in a new, different environment. This arises naturally when demonstrations are collected…

Machine Learning · Computer Science 2026-05-28 Guang-Yuan Hao , Lars van der Laan , Aurélien Bibaut , Nathan Kallus

Adversarial training is a standard defense against malicious input perturbations in security-critical machine-learning systems. Its main burden is structural: before every parameter update, the current model must first be attacked to find a…

Quantum Physics · Physics 2026-03-31 Yue Wang , Guangyi He , Liepeng Zhang , Lukas Gonon , Qi Zhao

Robust Reinforcement Learning tries to make predictions more robust to changes in the dynamics or rewards of the system. This problem is particularly important when the dynamics and rewards of the environment are estimated from the data. In…

Machine Learning · Computer Science 2022-06-15 Pierre Clavier , Stéphanie Allassonière , Erwan Le Pennec

In classical Q-learning, the objective is to maximize the sum of discounted rewards through iteratively using the Bellman equation as an update, in an attempt to estimate the action value function of the optimal policy. Conventionally, the…

Machine Learning · Computer Science 2019-06-25 Hadi S. Jomaa , Josif Grabocka , Lars Schmidt-Thieme

Offline reinforcement learning (RL) presents a promising approach for learning reinforced policies from offline datasets without the need for costly or unsafe interactions with the environment. However, datasets collected by humans in…

Machine Learning · Computer Science 2024-03-12 Rui Yang , Han Zhong , Jiawei Xu , Amy Zhang , Chongjie Zhang , Lei Han , Tong Zhang

We show that reinforcement learning with verifiable rewards (RLVR) can elicit strong mathematical reasoning in certain language models even with spurious rewards that have little, no, or even negative correlation with the correct answer.…

We develop a model selection approach to tackle reinforcement learning with adversarial corruption in both transition and reward. For finite-horizon tabular MDPs, without prior knowledge on the total amount of corruption, our algorithm…

Machine Learning · Computer Science 2024-12-31 Chen-Yu Wei , Christoph Dann , Julian Zimmert

Q-learning suffers from overestimation bias, because it approximates the maximum action value using the maximum estimated action value. Algorithms have been proposed to reduce overestimation bias, but we lack an understanding of how bias…

Machine Learning · Computer Science 2021-08-10 Qingfeng Lan , Yangchen Pan , Alona Fyshe , Martha White

One of the most natural approaches to reinforcement learning (RL) with function approximation is value iteration, which inductively generates approximations to the optimal value function by solving a sequence of regression problems. To…

Machine Learning · Computer Science 2024-06-19 Noah Golowich , Ankur Moitra

Temporal-difference learning (TD), coupled with neural networks, is among the most fundamental building blocks of deep reinforcement learning. However, due to the nonlinearity in value function approximation, such a coupling leads to…

Machine Learning · Computer Science 2020-04-16 Qi Cai , Zhuoran Yang , Jason D. Lee , Zhaoran Wang

Reward machines are an established tool for dealing with reinforcement learning problems in which rewards are sparse and depend on complex sequences of actions. However, existing algorithms for learning reward machines assume an overly…

Machine Learning · Computer Science 2025-10-20 Jan Corazza , Ivan Gavran , Daniel Neider

We study a simple model of algorithmic collusion in which Q-learning algorithms are designed in a strategic fashion. We let players (\textit{designers}) choose their exploration policy simultaneously prior to letting their algorithms…

Theoretical Economics · Economics 2024-09-13 Ivan Conjeaud

In this study, we leverage the deliberate and systematic fault-injection capabilities of an open-source benchmark suite to perform a series of experiments on state-of-the-art deep and robust reinforcement learning algorithms. We aim to…

Robotics · Computer Science 2022-10-28 Catherine R. Glossop , Jacopo Panerati , Amrit Krishnan , Zhaocong Yuan , Angela P. Schoellig

Ensuring safety via safety filters in real-world robotics presents significant challenges, particularly when the system dynamics is complex or unavailable. To handle this issue, learning-based safety filters recently gained popularity,…

Robotics · Computer Science 2024-12-02 Guo Ning Sue , Yogita Choudhary , Richard Desatnik , Carmel Majidi , John Dolan , Guanya Shi

Despite the fact that deep reinforcement learning (RL) has surpassed human-level performances in various tasks, it still has several fundamental challenges. First, most RL methods require intensive data from the exploration of the…

Machine Learning · Computer Science 2021-07-06 Zhe Xu , Bo Wu , Aditya Ojha , Daniel Neider , Ufuk Topcu

In recent work it is shown that Q-learning with linear function approximation is stable, in the sense of bounded parameter estimates, under the $(\varepsilon,\kappa)$-tamed Gibbs policy; $\kappa$ is inverse temperature, and $\varepsilon>0$…

Machine Learning · Computer Science 2026-02-09 Prashant Mehta , Sean Meyn

Safe reinforcement learning (RL) trains a policy to maximize the task reward while satisfying safety constraints. While prior works focus on the performance optimality, we find that the optimal solutions of many safe RL problems are not…

Machine Learning · Computer Science 2023-03-03 Zuxin Liu , Zijian Guo , Zhepeng Cen , Huan Zhang , Jie Tan , Bo Li , Ding Zhao

In high-stakes AI applications, even a single action can cause irreparable damage. However, nearly all of sequential decision-making theory assumes that all errors are recoverable (e.g., by bounding rewards). Standard bandit algorithms that…

Machine Learning · Computer Science 2026-04-14 Sarah Liaw , Benjamin Plaut

We study the exploration problem with approximate linear action-value functions in episodic reinforcement learning under the notion of low inherent Bellman error, a condition normally employed to show convergence of approximate value…

Machine Learning · Computer Science 2020-06-30 Andrea Zanette , Alessandro Lazaric , Mykel Kochenderfer , Emma Brunskill
‹ Prev 1 3 4 5 6 7 10 Next ›