AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

A New, Physics-Informed Continuous-Time Reinforcement Learning Algorithm with Performance Guarantees

Brent A. Wallace, Jennie Si

JMLR 2024 · Reinforcement learning › Policy gradient and actor-critic · approach: Model-based RL with guarantees

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use RCI instead of cFVIcontinuous-time RL control · Jet aircraft FPA loop1% settling time (s): 14.42 -> 7.17Improved
use RCI instead of cFVI at 25% modeling errorCT-RL · Jet aircraft FPA loop1% settling time (s): 15.20 -> 6.44Improved
use RCI instead of cFVICT-RL · Pendulum (nu=0%)average return: -8.85 -> -12.17Hurt
use RCI instead of cFVICT-RL · DDMR (nu=0%)average return: -4.12 -> -3.69Improved
use RCI instead of cFVICT-RL · Jet aircraft (nu=0%)average return: -20.50 -> -8.26Improved
use RCI instead of Nominal LQ (learn from nonlinear dynamics and data) (ablation)LQ design · Pendulum, Jet, DDMR with modeling errorpolicy optimality error: over 90% reduction in optimality errorImproved
use RCI instead of Nominal LQ at 25% modeling error (ablation)LQ design · Pendulumpolicy optimality error: 2.61 -> 0.13Improved
use RCI instead of Nominal LQ at 25% modeling error (ablation)LQ design · Pendulum y=theta1% settling time (s): 2.12 -> 1.18Improved
use RCI instead of Nominal LQ at 25% modeling error (ablation)LQ design · Jet aircraft FPA loopovershoot (%): 1.11 -> 0.69Improved
use RCI instead of cFVI at 25% modeling errorCT-RL · Jet aircraft FPA loopovershoot (%): 19.96 -> 0.69Improved
use RCI instead of cFVICT-RL · DDMR omega loop1% settling time (s): 12.81 -> 6.68Improved
use RCI instead of cFVICT-RL · DDMRtrajectory episodes required: RCI/FVI = 1/5,000,000Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy