English
Related papers

Related papers: Logit Dynamics in Softmax Policy Gradient Methods

200 papers

In this work, we consider policy-based methods for solving the reinforcement learning problem, and establish the sample complexity guarantees. A policy-based algorithm typically consists of an actor and a critic. We consider using various…

Machine Learning · Computer Science 2023-01-16 Zaiwei Chen , Siva Theja Maguluri

Safe Reinforcement learning (Safe RL) aims at learning optimal policies while staying safe. A popular solution to Safe RL is shielding, which uses a logical safety specification to prevent an RL agent from taking unsafe actions. However,…

Artificial Intelligence · Computer Science 2023-03-07 Wen-Chi Yang , Giuseppe Marra , Gavin Rens , Luc De Raedt

Many applications -- including power systems, robotics, and economics -- involve a dynamical system interacting with a stochastic and hard-to-model environment. We adopt a reinforcement learning approach to control such systems.…

Optimization and Control · Mathematics 2025-08-26 Abed AlRahman Al Makdah , Oliver Kosut , Lalitha Sankar , Shaofeng Zou

Soft robotic manipulators offer operational advantage due to their compliant and deformable structures. However, their inherently nonlinear dynamics presents substantial challenges. Traditional analytical methods often depend on simplifying…

Robotics · Computer Science 2024-10-28 Uljad Berdica , Matthew Jackson , Niccolò Enrico Veronese , Jakob Foerster , Perla Maiolino

Policy gradient methods are widely used in reinforcement learning. Yet, the nonconvexity of policy optimization poses significant challenges in understanding the global convergence of policy gradient methods. For a class of finite-horizon…

Optimization and Control · Mathematics 2026-03-10 Xin Chen , Yifan Hu , Minda Zhao

Risk-sensitive reinforcement learning (RL) is crucial for maintaining reliable performance in high-stakes applications. While traditional RL methods aim to learn a point estimate of the random cumulative cost, distributional RL (DRL) seeks…

Machine Learning · Computer Science 2025-02-03 Minheng Xiao , Xian Yu , Lei Ying

We introduce the first direct policy search algorithm which provably converges to the globally optimal $\textit{dynamic}$ filter for the classical problem of predicting the outputs of a linear dynamical system, given noisy, partial…

Optimization and Control · Mathematics 2022-03-01 Jack Umenberger , Max Simchowitz , Juan C. Perdomo , Kaiqing Zhang , Russ Tedrake

Linear dynamical systems that obey stochastic differential equations are canonical models. While optimal control of known systems has a rich literature, the problem is technically hard under model uncertainty and there are hardly any…

Systems and Control · Electrical Eng. & Systems 2023-06-09 Mohamad Kazem Shirani Faradonbeh , Mohamad Sadegh Shirani Faradonbeh

Calibration, which establishes the correlation between accuracy and model confidence, is important for LLM development. We design three off-the-shelf calibration methods based on self-consistency (Wang et al., 2022) for math reasoning…

Computation and Language · Computer Science 2024-03-18 Ante Wang , Linfeng Song , Ye Tian , Baolin Peng , Lifeng Jin , Haitao Mi , Jinsong Su , Dong Yu

In order to model risk aversion in reinforcement learning, an emerging line of research adapts familiar algorithms to optimize coherent risk functionals, a class that includes conditional value-at-risk (CVaR). Because optimizing the…

Machine Learning · Computer Science 2021-03-09 Audrey Huang , Liu Leqi , Zachary C. Lipton , Kamyar Azizzadenesheli

Policy gradient (PG) estimation becomes a challenge when we are not allowed to sample with the target policy but only have access to a dataset generated by some unknown behavior policy. Conventional methods for off-policy PG estimation…

Machine Learning · Statistics 2022-06-22 Chengzhuo Ni , Ruiqi Zhang , Xiang Ji , Xuezhou Zhang , Mengdi Wang

This is a short communication on a Lyapunov function argument for softmax in bandit problems. There are a number of excellent papers coming out using differential equations for policy gradient algorithms in reinforcement learning…

Machine Learning · Computer Science 2020-07-21 Neil Walton

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

A safe and trustworthy use of Large Language Models (LLMs) requires an accurate expression of confidence in their answers. We propose a novel Reinforcement Learning approach that allows to directly fine-tune LLMs to express calibrated…

Computation and Language · Computer Science 2026-03-03 David Bani-Harouni , Chantal Pellegrini , Paul Stangel , Ege Özsoy , Kamilia Zaripova , Nassir Navab , Matthias Keicher

When applying imitation learning techniques to fit a policy from expert demonstrations, one can take advantage of prior stability/robustness assumptions on the expert's policy and incorporate such control-theoretic prior knowledge…

Optimization and Control · Mathematics 2021-03-25 Aaron Havens , Bin Hu

This paper proposes a novel termination criterion, termed the advantage gap function, for finite state and action Markov decision processes (MDP) and reinforcement learning (RL). By incorporating this advantage gap function into the design…

Machine Learning · Computer Science 2026-03-24 Caleb Ju , Guanghui Lan

The softmax function combined with a cross-entropy loss is a principled approach to modeling probability distributions that has become ubiquitous in deep learning. The softmax function is defined by a lone hyperparameter, the temperature,…

Machine Learning · Computer Science 2020-10-16 Atish Agarwala , Jeffrey Pennington , Yann Dauphin , Sam Schoenholz

We study the training dynamics of gradient descent in a softmax self-attention layer trained to perform linear regression and show that a simple first-order optimization algorithm can converge to the globally optimal self-attention…

Machine Learning · Computer Science 2026-03-03 Gautam Goel , Mahdi Soltanolkotabi , Peter Bartlett

Recently, policy optimization for control purposes has received renewed attention due to the increasing interest in reinforcement learning. In this paper, we investigate the convergence of policy optimization for quadratic control of…

Optimization and Control · Mathematics 2020-02-12 Joao Paulo Jansch-Porto , Bin Hu , Geir Dullerud

Policy gradients methods apply to complex, poorly understood, control problems by performing stochastic gradient descent over a parameterized class of polices. Unfortunately, even for simple control problems solvable by standard dynamic…

Machine Learning · Computer Science 2022-06-22 Jalaj Bhandari , Daniel Russo