English
Related papers

Related papers: Sharp asymptotic theory for Q-learning with LDTZ l…

200 papers

We study reinforcement learning (RL) for a class of continuous-time linear-quadratic (LQ) control problems for diffusions, where states are scalar-valued and running control rewards are absent but volatilities of the state processes depend…

Machine Learning · Computer Science 2025-07-25 Yilie Huang , Yanwei Jia , Xun Yu Zhou

This work presents the first finite-time analysis for the last-iterate convergence of average-reward $Q$-learning with an asynchronous implementation. A key feature of the algorithm we study is the use of adaptive stepsizes, which serve as…

Machine Learning · Computer Science 2026-04-07 Zaiwei Chen , Phalguni Nanda

Q-learning is a promising method for solving optimal control problems for uncertain systems without the explicit need for system identification. However, approaches for continuous-time Q-learning have limited provable safety guarantees,…

Systems and Control · Electrical Eng. & Systems 2024-01-30 Soutrik Bandyopadhyay , Shubhendu Bhasin

In this paper, we establish a theoretical comparison between the asymptotic mean-squared error of Double Q-learning and Q-learning. Our result builds upon an analysis for linear stochastic approximation based on Lyapunov equations and…

Machine Learning · Computer Science 2022-06-16 Wentao Weng , Harsh Gupta , Niao He , Lei Ying , R. Srikant

Stochastic Approximation (SA) is a widely used algorithmic approach in various fields, including optimization and reinforcement learning (RL). Among RL algorithms, Q-learning is particularly popular due to its empirical success. In this…

Machine Learning · Statistics 2024-01-26 Yixuan Zhang , Qiaomin Xie

Q-learning, which seeks to learn the optimal Q-function of a Markov decision process (MDP) in a model-free fashion, lies at the heart of reinforcement learning. When it comes to the synchronous setting (such that independent samples for all…

Machine Learning · Statistics 2025-03-18 Gen Li , Changxiao Cai , Yuxin Chen , Yuting Wei , Yuejie Chi

Distributionally robust reinforcement learning (DRRL) focuses on designing policies that achieve good performance under model uncertainties. The goal is to maximize the worst-case long-term discounted reward, where the data for RL comes…

Machine Learning · Computer Science 2026-03-17 Saptarshi Mandal , Yashaswini Murthy , R. Srikant

This paper studies an infinite horizon optimal control problem for discrete-time linear systems and quadratic criteria, both with random parameters which are independent and identically distributed with respect to time. A classical approach…

Optimization and Control · Mathematics 2020-11-11 Kai Du , Qingxin Meng , Fu Zhang

Deep Q-Learning (DQL), a family of temporal difference algorithms for control, employs three techniques collectively known as the `deadly triad' in reinforcement learning: bootstrapping, off-policy learning, and function approximation.…

Machine Learning · Computer Science 2019-03-22 Joshua Achiam , Ethan Knight , Pieter Abbeel

We analyse quantile temporal-difference learning (QTD), a distributional reinforcement learning algorithm that has proven to be a key component in several successful large-scale applications of reinforcement learning. Despite these…

The current paper studies the problem of agnostic $Q$-learning with function approximation in deterministic systems where the optimal $Q$-function is approximable by a function in the class $\mathcal{F}$ with approximation error $\delta \ge…

Machine Learning · Computer Science 2020-02-18 Simon S. Du , Jason D. Lee , Gaurav Mahajan , Ruosong Wang

In the Economic Nonlinear Model Predictive (ENMPC) context, closed-loop stability relates to the existence of a storage function satisfying a dissipation inequality. Finding the storage function is in general -- for nonlinear dynamics and…

Systems and Control · Electrical Eng. & Systems 2021-10-26 Arash Bahari Kordabad , Sebastien Gros

$Q$-learning with function approximation is one of the most popular methods in reinforcement learning. Though the idea of using function approximation was proposed at least 60 years ago, even in the simplest setup, i.e, approximating…

Machine Learning · Computer Science 2019-11-05 Simon S. Du , Yuping Luo , Ruosong Wang , Hanrui Zhang

Learning rate schedules used in practice bear little resemblance to those recommended by theory. We close much of this theory/practice gap, and as a consequence are able to derive new problem-adaptive learning rate schedules. Our main…

Machine Learning · Computer Science 2024-10-31 Aaron Defazio , Ashok Cutkosky , Harsh Mehta , Konstantin Mishchenko

In this paper, we introduce a unified framework for analyzing a large family of Q-learning algorithms, based on switching system perspectives and ODE-based stochastic approximation. We show that the nonlinear ODE models associated with…

Optimization and Control · Mathematics 2021-02-18 Donghwan Lee , Niao He

In an episodic Markov Decision Process (MDP) problem, an online algorithm chooses from a set of actions in a sequence of $H$ trials, where $H$ is the episode length, in order to maximize the total payoff of the chosen actions. Q-learning,…

Machine Learning · Computer Science 2019-07-11 Xu Zhu

Large-scale multi-agent systems are often deployed across wide geographic areas, where agents interact with heterogeneous environments. There is an emerging interest in understanding the role of heterogeneity in the performance of the…

Machine Learning · Computer Science 2026-05-18 Leo Muxing Wang , Pengkun Yang , Lili Su

Two of the leading approaches for model-free reinforcement learning are policy gradient methods and $Q$-learning methods. $Q$-learning methods can be effective and sample-efficient when they work, however, it is not well-understood why they…

Machine Learning · Computer Science 2018-10-16 John Schulman , Xi Chen , Pieter Abbeel

We conjecture that the inherent difference in generalisation between adaptive and non-adaptive gradient methods in deep learning stems from the increased estimation noise in the flattest directions of the true loss surface. We demonstrate…

Machine Learning · Statistics 2022-03-17 Diego Granziol , Nicholas Baskerville

We test whether the optimal learning-rate schedule depends on bit-width during from-initialisation quantisation-aware training (QAT) for sub-100M decoder language models. A 720-run factorial grid (Phase 2) over bit-width x warmdown fraction…

Machine Learning · Computer Science 2026-05-26 Christian Brandt Thomassen