تقرير
A General Control-Theoretic Approach for Reinforcement Learning: Theory and Algorithms
العنوان: | A General Control-Theoretic Approach for Reinforcement Learning: Theory and Algorithms |
---|---|
المؤلفون: | Chen, Weiqin, Squillante, Mark S., Wu, Chai Wah, Paternain, Santiago |
سنة النشر: | 2024 |
المجموعة: | Computer Science Statistics |
مصطلحات موضوعية: | Computer Science - Machine Learning, Statistics - Methodology |
الوصف: | We devise a control-theoretic reinforcement learning approach to support direct learning of the optimal policy. We establish various theoretical properties of our approach, such as convergence and optimality of our control-theoretic operator, a new control-policy-parameter gradient ascent theorem, and a specific gradient ascent algorithm based on this theorem. As a representative example, we adapt our approach to a particular control-theoretic framework and empirically evaluate its performance on several classical reinforcement learning tasks, demonstrating significant improvements in solution quality, sample complexity, and running time of our control-theoretic approach over state-of-the-art baseline methods. |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2406.14753 |
رقم الأكسشن: | edsarx.2406.14753 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |