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
Maak v1.0.0 :: infinitely extensible command runner, build sys in Lisp
0Lobsters
Not Safe, Not Creative: 123456 Is Still the World’s Most Common Password
0CNET News

GTA 6 dev says removing cops from the player’s radar created ‘some really memorable and epic chases and escapes’ — ‘The player needs to now use the awareness/threat indicators to help them evade, or c
0TechRadar

Apple Card doubles Daily Cash rewards for new customers in limited-time offer
09to5Mac

SmartWings Black Friday sale: Save up to 15% on custom motorized shades
09to5Mac

Google AI Ultra upgrade adds YouTube Premium Family
09to5Google