English
Related papers

Related papers: Leverage the Average: an Analysis of KL Regulariza…

200 papers

Reinforcement learning (RL)-based post-training often improves the reasoning performance of large language models (LLMs) beyond the training domain, while supervised fine-tuning (SFT) frequently leads to general capabilities forgetting.…

Computation and Language · Computer Science 2026-04-29 Dan Shi , Zhuowen Han , Simon Ostermann , Renren Jin , Josef van Genabith , Deyi Xiong

Maximum regularized likelihood estimators (MRLEs) are arguably the most established class of estimators in high-dimensional statistics. In this paper, we derive guarantees for MRLEs in Kullback-Leibler divergence, a general measure of…

Machine Learning · Statistics 2018-10-18 Rui Zhuang , Johannes Lederer

Language model (LM) alignment improves model outputs to reflect human preferences while preserving the capabilities of the base model. The most common alignment approaches are (i) reinforcement learning, which maximizes the expected reward…

Machine Learning · Computer Science 2026-05-11 Lucas Monteiro Paes , Natalie Mackraz , Barry-John Theobald , Federico Danieli

Statistical leverage scores emerged as a fundamental tool for matrix sketching and column sampling with applications to low rank approximation, regression, random feature learning and quadrature. Yet, the very nature of this quantity is…

Machine Learning · Computer Science 2018-11-22 Edouard Pauwels , Francis Bach , Jean-Philippe Vert

Deep reinforcement learning (RL) algorithms are predominantly evaluated by comparing their relative performance on a large suite of tasks. Most published results on deep RL benchmarks compare point estimates of aggregate performance such as…

Machine Learning · Computer Science 2022-01-06 Rishabh Agarwal , Max Schwarzer , Pablo Samuel Castro , Aaron Courville , Marc G. Bellemare

Reinforcement learning (RL) algorithms allow agents to learn skills and strategies to perform complex tasks without detailed instructions or expensive labelled training examples. That is, RL agents can learn, as we learn. Given the…

Machine Learning · Computer Science 2019-01-25 Jung Hoon Lee

We provide a framework for accelerating reinforcement learning (RL) algorithms by heuristics constructed from domain knowledge or offline data. Tabula rasa RL algorithms require environment interactions or computation that scales with the…

Machine Learning · Computer Science 2021-11-23 Ching-An Cheng , Andrey Kolobov , Adith Swaminathan

We introduce a framework for Continual Learning (CL) based on Bayesian inference over the function space rather than the parameters of a deep neural network. This method, referred to as functional regularisation for Continual Learning,…

Reinforcement Learning (RL) has emerged as a crucial paradigm for unlocking the advanced reasoning capabilities of Large Language Models (LLMs), encompassing frameworks like RLHF and RLAIF. Regardless of the specific optimization algorithm…

Machine Learning · Computer Science 2026-05-11 Rui Zhu , Weiheng Bai , Qiushi Wu , Yang Ren , Haixu Tang , Yuchu Liu

Reinforcement learning (RL) has been widely applied to sequential decision making, where interpretability and performance are both critical for practical adoption. Current approaches typically focus on performance and rely on post hoc…

Machine Learning · Computer Science 2025-10-07 Qianxin Yi , Shao-Bo Lin , Jun Fan , Yao Wang

Speculative decoding accelerates autoregressive large language model (LLM) inference by using a lightweight draft model to propose candidate tokens that are then verified in parallel by the target model. The speedup is significantly…

Machine Learning · Computer Science 2026-03-02 Alexander Samarin , Sergei Krutikov , Anton Shevtsov , Sergei Skvortsov , Filipp Fisin , Alexander Golubev

Kullback--Leibler (KL) divergence is a fundamental measure of the dissimilarity between two probability distributions, but it can become unstable in high-dimensional settings due to its sensitivity to mismatches in distributional support.…

Information Theory · Computer Science 2025-02-03 Yifeng Peng , Dantong Li , Xinyi Li , Zhiding Liang , Yongshan Ding , Ying Wang

Reinforcement Learning (RL) has emerged as a highly effective technique for addressing various scientific and applied problems. Despite its success, certain complex tasks remain challenging to be addressed solely with a single model and…

Machine Learning · Computer Science 2023-12-14 Yanjie Song , P. N. Suganthan , Witold Pedrycz , Junwei Ou , Yongming He , Yingwu Chen , Yutong Wu

We propose a general, yet simple patch that can be applied to existing regularization-based continual learning methods called classifier-projection regularization (CPR). Inspired by both recent results on neural networks with wide local…

Machine Learning · Computer Science 2021-04-20 Sungmin Cha , Hsiang Hsu , Taebaek Hwang , Flavio P. Calmon , Taesup Moon

Logistic regression is commonly used for modeling dichotomous outcomes. In the classical setting, where the number of observations is much larger than the number of parameters, properties of the maximum likelihood estimator in logistic…

Machine Learning · Statistics 2019-11-14 Fariborz Salehi , Ehsan Abbasi , Babak Hassibi

Group Relative Policy Optimization (GRPO) is widely used for critic-free Large Language Model (LLM) post-training, but its KL regularization is usually implemented as a local loss-side token penalty. We show that this misses the…

Machine Learning · Computer Science 2026-05-26 Jiarui Yao , Ruida Wang , Hao Bai , Tong Zhang

The ability to achieve long-term goals is a key challenge in the current development of large language models (LLMs). To address this, pre-trained LLMs can be fine-tuned with reinforcement learning (RL) to explore solutions that optimize a…

Computation and Language · Computer Science 2025-02-11 Jean Vassoyan , Nathanaël Beau , Roman Plaud

Designing optimal controllers continues to be challenging as systems are becoming complex and are inherently nonlinear. The principal advantage of reinforcement learning (RL) is its ability to learn from the interaction with the environment…

Machine Learning · Computer Science 2018-10-05 Savinay Nagendra , Nikhil Podila , Rashmi Ugarakhod , Koshy George

Reinforcement learning with verifiable rewards (RLVR) has become a trending paradigm for training reasoning large language models (LLMs). However, due to the autoregressive decoding nature of LLMs, the rollout process becomes the efficiency…

Machine Learning · Computer Science 2026-02-17 Yuhang Li , Reena Elangovan , Xin Dong , Priyadarshini Panda , Brucek Khailany

Modern deep reinforcement learning (RL) algorithms are motivated by either the generalised policy iteration (GPI) or trust-region learning (TRL) frameworks. However, algorithms that strictly respect these theoretical frameworks have proven…

Machine Learning · Computer Science 2024-11-21 Jakub Grudzien Kuba , Christian Schroeder de Witt , Jakob Foerster
‹ Prev 1 8 9 10 Next ›