AI Paper Atlas

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

Value Iteration in Continuous Actions, States and Time

Michael Lutter, Shie Mannor, Jan Peters, Dieter Fox, Animesh Garg

ICML 2021 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory · 8 citations (OpenAlex)

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 offline cFVI instead of the strongest listed deep RL baselineSwing-up control · Simulated Pendulumreward: -030.6 ± 001.4 -> -030.5 ± 000.8Improved
Use offline cFVI instead of the strongest listed deep RL baselineSwing-up control · Simulated Cartpolereward: -024.2 ± 001.4 -> -024.2 ± 002.1No clear effect
Use offline cFVI instead of the strongest listed deep RL baselineSwing-up control · Simulated Furuta Pendulumreward: -028.1 ± 002.0 -> -027.7 ± 001.6Improved
Use offline cFVI instead of PPO with uniform initial states and domain randomizationSwing-up control · Physical Cartpolereward: -143.8 ± 016.1 -> -143.7 ± 210.4No clear effect
Use offline cFVI instead of PPO with uniform initial states and domain randomizationSwing-up control · Physical Furuta Pendulumreward: -080.6 ± 010.8 -> -082.1 ± 007.6No clear effect
Use RTDP cFVI instead of offline DP cFVIcFVI · Physical Furuta Pendulumreward: -082.1 ± 007.6 -> -1009.9 ± 004.5Hurt
Increase the n-step horizon using larger λ (ablation)DP cFVI · Swing-up control tasksconvergence rateImproved
Use an extremely long value-target horizon (ablation)DP cFVI · Furuta Pendulum and Cartpolelearning stabilityHurt
Increase the value-target horizon from λ = 0.01 to λ = 0.99 (ablation)cFVI · Pendulumforward rollout time: 0.4s -> 56.4sHurt
Use a locally quadratic value function rather than an MLP (ablation)cFVI · Control taskslearning speedImproved
Use a standard network architecture without the locally quadratic structure or feature transform (ablation)cFVI · Cartpolepolicy qualityHurt
Add the feature transform to the locally quadratic value function (ablation)cFVI · Cartpolepolicy qualityImproved

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