AI & ML
First-Order Model-Free Policy Gradient for Undiscounted LQR with Global Optimality and Convergence Guarantees
arXiv cs.SY0 views
arXiv:2611.21985v1 Announce Type: new Abstract: We provide a first-order model-free policy gradient (PG) framework for learning undiscounted infinite-horizon linear quadratic regulator (LQR) policies with global optimality and convergence guarantees. We first construct a stochastic wrapper policy class parameterized by the LQR gain to enable exploration around a deterministic linear control law. We analyze the additional cost induced by exploration over infinite-horizon, and establish a weak gradient dominance condition that characterizes global optimality under exploration. We then construct a model-free first-order PG algorithm using sampled trajectory information, without identifying the dynamics or approximating a value function. Finally, we establish finite-time convergence guarantees, providing the sufficient number of samples required for near-global optimality with our proposed PG algorithms.
Source:https://arxiv.org/abs/2611.21985PreviousEmerging Vertical Power Delivery in AI Data Centers: Challenges and SolutionsNextSafe Reinforcement Learning with Hard Constraints: Policy Gradient Methods with Intersample Guarantees
Related posts
How Far Behind Are Open Weights, Measured
0dev.to
Claude was trained on protein sequences at scale
0Hacker News
Pretraining vs post-training: one apple problem, four learning signals
0dev.to Beginners
Bitcoin Recovery Expert USA: Professional Crypto Asset Recovery with META TECH RETRIEVE PRO
0dev.to Beginners
Hugging Face Switches Its Inference Providers to OpenRouter Routing
0dev.to OpenSource
Still confused about Git vs GitHub?
0dev.to Git