Newest papers

Continuous-Time Policy Gradient from Discrete Data: Convergence Analysis and Model-Free Algorithms

Martin Hernandez, Xinyan Qing, Mo Zhou, Yuhua Zhu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28580 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Pembelajaran penguatan data berkesinambungan mencari kebijakan optimal di bawah dinamika yang tidak diketahui yang didorong oleh persamaan diferensial stochastic menggunakan pengamatan yang dikumpulkan pada waktu yang terpisah. Kami mengembangkan kerangka kebijakan gradient yang menggabungkan analisis konvergensi yang ketat dengan belajar dari data yang terpisah. Kami pertama-tama menetapkan lemma perbedaan kinerja dan mendistribusikan rumus gradient kebijakan untuk kebijakan acak dan parameter. Kami kemudian membuktikan konvergensi eksponensial dari aliran kebijakan yang diatur entropy ke optimum yang diatur. Untuk pembaruan parameter yang terpisah, kami mendistribusikan ikatan konvergensi yang secara eksplisit memisahkan kontribusi kesalahan kritis, kesalahan kebijakan-regresi, dan ukuran langkah pembelajaran. Untuk memungkinkan belajar di bawah dinamika

Machine translation (Meta M2M100); the English original follows.

Abstract

Continuous-time reinforcement learning seeks an optimal policy under unknown dynamics driven by stochastic differential equations using observations collected at discrete times. We develop a policy gradient framework that combines rigorous convergence analysis with learning from discrete data. We first establish a performance difference lemma and derive policy gradient formulas for randomized and parametrized policies. We then prove exponential convergence of an entropy-regularized policy flow to the regularized optimum. For discrete parametrized updates, we derive a convergence bound that explicitly separates the contributions of critic error, policy-regression error, and the learning step size. To enable learning under unknown dynamics, we integrate this framework with PhiBE, which constructs a continuous-time surrogate generator from discrete transition data. The resulting on-policy and off-policy model-free actor-critic algorithms preserve the continuous-time structure of the control problem. Numerical experiments on an entropy-regularized linear-quadratic regulator and a nonlinear wheeled robot demonstrate the performance and stability of the proposed methods with both known and data-driven dynamics.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Convergence (economics), Reinforcement learning, Mathematics, Lemma (botany), Stability (learning theory), Discrete time and continuous time

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange