English
Related papers

Related papers: You May Not Need Ratio Clipping in PPO

200 papers

We propose a new sample-efficient methodology, called Supervised Policy Update (SPU), for deep reinforcement learning. Starting with data generated by the current policy, SPU formulates and solves a constrained optimization problem in the…

Machine Learning · Computer Science 2018-12-27 Quan Vuong , Yiming Zhang , Keith W. Ross

We study differentially private (DP) algorithms for stochastic convex optimization (SCO). In this problem the goal is to approximately minimize the population loss given i.i.d. samples from a distribution over convex and Lipschitz loss…

Machine Learning · Computer Science 2019-08-28 Raef Bassily , Vitaly Feldman , Kunal Talwar , Abhradeep Thakurta

Stochastic computer simulations enable users to gain new insights into complex physical systems. Optimization is a common problem in this context: users seek to find model inputs that maximize the expected value of an objective function.…

Optimization and Control · Mathematics 2018-09-13 Atiye Alaeddini , Daniel J. Klein

Policy gradient algorithms have driven many recent advancements in language model reasoning. An appealing property is their ability to learn from exploration on their own trajectories, a process crucial for fostering diverse and creative…

Machine Learning · Computer Science 2026-03-13 Aleksei Petrenko , Ben Lipkin , Kevin Chen , Erik Wijmans , Marco Cusumano-Towner , Raja Giryes , Philipp Krähenbühl

Policy optimization is among the most popular and successful reinforcement learning algorithms, and there is increasing interest in understanding its theoretical guarantees. In this work, we initiate the study of policy optimization for the…

Machine Learning · Computer Science 2022-02-08 Liyu Chen , Haipeng Luo , Aviv Rosenberg

Existing post-training techniques are broadly categorized into supervised fine-tuning (SFT) and reinforcement learning (RL) methods; the former is stable during training but suffers from limited generalization, while the latter, despite its…

Computer Vision and Pattern Recognition · Computer Science 2025-07-02 Daoan Zhang , Guangchen Lan , Dong-Jun Han , Wenlin Yao , Xiaoman Pan , Hongming Zhang , Mingxiao Li , Pengcheng Chen , Yu Dong , Christopher Brinton , Jiebo Luo

Traditional RL algorithms like Proximal Policy Optimization (PPO) typically train on the entire rollout buffer, operating under the assumption that all generated episodes provide a beneficial optimization signal. However, these episodes…

Machine Learning · Computer Science 2026-04-03 Dong Shu , Denghui Zhang , Jessica Hullman

Most provably-efficient learning algorithms introduce optimism about poorly-understood states and actions to encourage exploration. We study an alternative approach for efficient exploration, posterior sampling for reinforcement learning…

Machine Learning · Statistics 2013-12-30 Ian Osband , Daniel Russo , Benjamin Van Roy

Existing reinforcement learning approaches for Large Language Models typically perform policy optimization at the granularity of individual tokens or entire response sequences. However, such formulations often misalign with the natural…

Artificial Intelligence · Computer Science 2026-05-08 Lei Gao , Zhuoming Li , Mengxi Jia , Jiakang Yuan , Hongbo Sun , Hao Sun , Xuelong Li

Early stopping is a well known approach to reduce the time complexity for performing training and model selection of large scale learning machines. On the other hand, memory/space (rather than time) complexity is the main constraint in many…

Machine Learning · Statistics 2018-02-02 Tomas Angles , Raffaello Camoriano , Alessandro Rudi , Lorenzo Rosasco

For many applications of reinforcement learning it can be more convenient to specify both a reward function and constraints, rather than trying to design behavior through the reward function. For example, systems that physically interact…

Machine Learning · Computer Science 2017-05-31 Joshua Achiam , David Held , Aviv Tamar , Pieter Abbeel

Policy gradient methods usually rely on entropy regularization to prevent premature convergence. However, maximizing entropy indiscriminately pushes the policy towards a uniform distribution, often overriding the reward signal if not…

Machine Learning · Computer Science 2026-03-06 Luca Serfilippi , Giorgio Franceschelli , Antonio Corradi , Mirco Musolesi

Reinforcement learning (RL)-based fine-tuning has emerged as a powerful approach for aligning diffusion models with black-box objectives. Proximal policy optimization (PPO) is a popular choice of method for policy optimization. While…

Group Relative Policy Optimisation (GRPO) enhances large language models by estimating advantages across a group of sampled trajectories. However, mapping these trajectory-level advantages to policy updates requires aggregating token-level…

Policy optimization for large language models often suffers from sparse reward signals in multi-step reasoning tasks. Critic-free methods like GRPO assign a single normalized outcome reward to all tokens, providing limited guidance for…

Machine Learning · Computer Science 2026-02-04 Ruiyi Ding , Yongxuan Lv , Xianhui Meng , Jiahe Song , Chao Wang , Chen Jiang , Yuan Cheng

In this paper, we propose a sparse equity portfolio optimization (SEPO) based on the mean-variance portfolio selection model. Aimed at minimizing transaction cost by avoiding small investments, this new model includes $\ell_0$-norm…

Optimization and Control · Mathematics 2021-09-14 Hong Seng Sim , Wendy Shin Yie Ling , Wah June Leong , Chuei Yee Chen

Solving tasks in Reinforcement Learning is no easy feat. As the goal of the agent is to maximize the accumulated reward, it often learns to exploit loopholes and misspecifications in the reward signal resulting in unwanted behavior. While…

Machine Learning · Computer Science 2018-12-27 Chen Tessler , Daniel J. Mankowitz , Shie Mannor

Preference optimization (PO) is indispensable for large language models (LLMs), with methods such as direct preference optimization (DPO) and proximal policy optimization (PPO) achieving great success. A common belief is that DPO is…

Machine Learning · Computer Science 2026-05-18 Yue Wang , Qizhou Wang , Zizhuo Zhang , Gang Niu , Bo Han , Masashi Sugiyama

Large language models (LLMs) have achieved great success across diverse tasks, and fine-tuning is sometimes needed to further enhance generation quality. Most existing methods rely on human supervision or parameter retraining, both of which…

Computation and Language · Computer Science 2025-05-27 Zhen-Yu Zhang , Jiandong Zhang , Huaxiu Yao , Gang Niu , Masashi Sugiyama

There has been a growing interest in off-policy evaluation in the literature such as recommender systems and personalized medicine. We have so far seen significant progress in developing estimators aimed at accurately estimating the…

Machine Learning · Computer Science 2024-04-24 Yuta Saito , Masahiro Nomura