Continuous-Time RL

RL: Reinforcement Learning

Ask about this topic

Overview

Continuous-time reinforcement learning studies how to choose actions and their timing when states evolve between decisions, as in physical control, finance, and medical treatment. Methods include actor–critic and policy-gradient algorithms, Q-learning, model-based approaches that estimate diffusion generators, and entropy regularization for exploration and robustness. Researchers apply these methods to stochastic differential equations, jump processes, mean-field systems, and optimal stopping, evaluating them on control, pricing, and treatment tasks. The field has growing theoretical guarantees for convergence and regret, but practical learning still depends on approximations to dynamics and value functions, with open challenges in data efficiency, safety, and path-dependent systems.

Latest papers

CardsList

34 papers4 papers in the last four weeks, up 33% on the four weeks before. 0.0% of all new papers.

  1. Learning When to Act: Interval-Aware Reinforcement Learning with Predictive Temporal Structure

    Mar 23, 2026Davide Di GioiaContinuous-Time RLCredit Assignment in RL

    University College London

  2. Safe Continuous-time Multi-Agent Reinforcement Learning via Epigraph Form

    Feb 19, 2026Xuefeng Wang, Lei Zhang, Henglin Pu +2Multi-Agent Reinforcement LearningContinuous-Time RL

    Purdue University

  3. Decoupled Continuous-Time Reinforcement Learning via Hamiltonian Flow

    Feb 16, 2026Minh NguyenReinforcement LearningContinuous-Time RL

    Google · University of Texas at Austin

  4. Infinitesimal Operator World Models: Finite-Sample Guarantees for Offline Reinforcement Learning

    Nov 13, 2025Nicolas Hoischen, Petar Bevanda, Max Beier +3Offline RLModel-Based RL

    TU Munich · ShanghaiTech University