English
Related papers

Related papers: Model-free Low-Rank Reinforcement Learning via Lev…

200 papers

Policy iteration (PI) is a recursive process of policy evaluation and improvement for solving an optimal decision-making/control problem, or in other words, a reinforcement learning (RL) problem. PI has also served as the fundamental for…

Artificial Intelligence · Computer Science 2021-04-06 Jaeyoung Lee , Richard S. Sutton

Scaling critic capacity is a promising direction for improving off-policy reinforcement learning (RL). However, recent work shows that larger critics are prone to overfitting and instability in replay-based bootstrapped training. In this…

Machine Learning · Computer Science 2026-05-08 Yuan Zhuang , Yuexin Bian , Sihong He , Jie Feng , Qing Su , Songyang Han , Jonathan Petit , Shihao Ji , Yuanyuan Shi , Fei Miao

Low Rank Adaptation (LoRA) is a popular Parameter Efficient Fine Tuning (PEFT) method that effectively adapts large pre-trained models for downstream tasks. LoRA parameterizes model updates using low-rank matrices at each layer,…

Computation and Language · Computer Science 2025-02-04 Ignacio Hounie , Charilaos Kanatsoulis , Arnuv Tandon , Alejandro Ribeiro

Fine-tuning large language models (LLMs) is crucial for improving their performance on downstream tasks, but full-parameter fine-tuning (Full-FT) is computationally expensive and memory-intensive. Parameter-efficient fine-tuning (PEFT)…

Computation and Language · Computer Science 2026-05-12 Longteng Zhang , Lin Zhang , Shaohuai Shi , Xiaowen Chu , Bo Li

Policy evaluation is a core component of many reinforcement learning (RL) algorithms and a critical tool for ensuring safe deployment of RL policies. However, existing policy evaluation methods often suffer from high variance or bias. To…

Artificial Intelligence · Computer Science 2026-03-23 Shripad Vilasrao Deshmukh , Will Schwarzer , Scott Niekum

Low-Rank Adaptation (LoRA) lowers the computational and memory overhead of fine-tuning large models by updating a low-dimensional subspace of the pre-trained weight matrix. Albeit efficient, LoRA exhibits suboptimal convergence and…

Machine Learning · Computer Science 2026-02-25 Yilang Zhang , Bingcong Li , Georgios B. Giannakis

We present an algorithm for local, regularized, policy improvement in reinforcement learning (RL) that allows us to formulate model-based and model-free variants in a single framework. Our algorithm can be interpreted as a natural extension…

We study matrix estimation problems arising in reinforcement learning (RL) with low-rank structure. In low-rank bandits, the matrix to be recovered specifies the expected arm rewards, and for low-rank Markov Decision Processes (MDPs), it…

Machine Learning · Computer Science 2023-10-31 Stefan Stojanovic , Yassir Jedra , Alexandre Proutiere

Probabilistic learning to rank (LTR) has been the dominating approach for optimizing the ranking metric, but cannot maximize long-term rewards. Reinforcement learning models have been proposed to maximize user long-term rewards by…

Machine Learning · Computer Science 2024-01-18 Teng Xiao , Suhang Wang

The practicality of reinforcement learning algorithms has been limited due to poor scaling with respect to the problem size, as the sample complexity of learning an $\epsilon$-optimal policy is $\tilde{\Omega}\left(|S||A|H^3 /…

Machine Learning · Computer Science 2023-06-12 Tyler Sam , Yudong Chen , Christina Lee Yu

In this paper we consider the problem of learning an $\epsilon$-optimal policy for a discounted Markov Decision Process (MDP). Given an MDP with $S$ states, $A$ actions, the discount factor $\gamma \in (0,1)$, and an approximation threshold…

Machine Learning · Computer Science 2020-12-25 Zihan Zhang , Yuan Zhou , Xiangyang Ji

Low-Rank Adaptation (LoRA) is a parameter-efficient technique for rapidly fine-tuning foundation models. In standard LoRA training dynamics, models tend to quickly converge to a local optimum near the initialization. However, this local…

Machine Learning · Computer Science 2024-10-31 Zhan Zhuang , Xiequn Wang , Yulong Zhang , Wei Li , Yu Zhang , Ying Wei

Fine-tuning large-scale pre-trained models is prohibitively expensive in terms of computation and memory costs. Low-Rank Adaptation (LoRA), a popular Parameter-Efficient Fine-Tuning (PEFT) method, offers an efficient solution by optimizing…

Machine Learning · Computer Science 2025-05-27 Tao Li , Zhengbao He , Yujun Li , Yasheng Wang , Lifeng Shang , Xiaolin Huang

Many problems in computer vision and recommender systems involve low-rank matrices. In this work, we study the problem of finding the maximum entry of a stochastic low-rank matrix from sequential observations. At each step, a learning agent…

Machine Learning · Computer Science 2017-12-14 Branislav Kveton , Csaba Szepesvari , Anup Rao , Zheng Wen , Yasin Abbasi-Yadkori , S. Muthukrishnan

Reinforcement learning (RL) approaches for Large Language Models (LLMs) frequently use on-policy algorithms, such as PPO or GRPO. However, policy lag from distributed training architectures and differences between the training and inference…

Machine Learning · Computer Science 2026-03-03 Daniel Ritter , Owen Oertell , Bradley Guo , Jonathan Chang , Kianté Brantley , Wen Sun

Policy evaluation is an important instrument for the comparison of different algorithms in Reinforcement Learning (RL). However, even a precise knowledge of the value function $V^{\pi}$ corresponding to a policy $\pi$ does not provide…

Machine Learning · Computer Science 2026-01-21 Denis Belomestny , Ilya Levin , Alexey Naumov , Sergey Samsonov

Multi-agent reinforcement learning (MARL) often relies on \emph{parameter sharing (PS)} to scale efficiently. However, purely shared policies can stifle each agent's unique specialization, reducing overall performance in heterogeneous…

Multiagent Systems · Computer Science 2025-02-11 Beining Zhang , Aditya Kapoor , Mingfei Sun

LoRA has become a universal Parameter-Efficient Fine-Tuning (PEFT) technique that equips Large Language Models (LLMs) to adapt quickly to new tasks. However, when these models are scaled up, even the latest LoRA variants still introduce…

Computation and Language · Computer Science 2026-02-25 Xindian Ma , Rundong Kong , Peng Zhang , Ruoxiang Huang , Yongyu Jiang

Multi-step greedy policies have been extensively used in model-based reinforcement learning (RL), both when a model of the environment is available (e.g.,~in the game of Go) and when it is learned. In this paper, we explore their benefits…

Machine Learning · Computer Science 2020-07-14 Manan Tomar , Yonathan Efroni , Mohammad Ghavamzadeh

We propose a policy improvement algorithm for Reinforcement Learning (RL) which is called Rerouted Behavior Improvement (RBI). RBI is designed to take into account the evaluation errors of the Q-function. Such errors are common in RL when…

Machine Learning · Computer Science 2019-07-12 Elad Sarafian , Aviv Tamar , Sarit Kraus