AI Paper Atlas

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

Deterministic Value-Policy Gradients

Qingpeng Cai, Ling Pan, Pingzhong Tang

AAAI 2020 · Reinforcement learning › Policy gradient and actor-critic · approach: Variance-reduced policy gradients · 1 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
Ensemble model-based deterministic value gradients with model-free deterministic policy gradients using geometric-decay weightsDVPG · MuJoCo continuous-control benchmarksepisode rewardImproved
Use infinite-horizon value gradients rather than finite-horizon gradientsDVG · HumanoidStandup-v2episode rewardImproved
Use model-based value-gradient rollouts instead of no model rolloutDVG · Hopper-v2sample efficiencyImproved
Use one model rollout step rather than DDPG's zero-step model-free gradientDVG · Hopper-v2episode rewardImproved
Increase model rollout steps to 5 (ablation)DVG · Hopper-v2training stabilityHurt
Use a discount factor of 0.95 rather than 0.8 or 0.6 (ablation)DVG with 2 rollout steps · InvertedPendulum-v2episode reward and stabilityImproved
Use discount factors 0.85 or 0.99 instead of 0.95 (ablation)DVG with 2 rollout steps · InvertedPendulum-v2episode reward and stabilityNo clear effect
Use discount factors 0.8 or 0.6 instead of 0.95 (ablation)DVG with 2 rollout steps · InvertedPendulum-v2episode rewardHurt
Increase the bootstrapping weight λ (ablation)DVPG with 4 rollout steps · MuJoCo continuous-control benchmarkssample efficiency and episode rewardHurt
Set the bootstrapping weight λ to 0.1 (ablation)DVPG with 4 rollout steps · MuJoCo continuous-control benchmarkssample efficiency and episode rewardImproved
Use 1 rollout step rather than other tested rollout-step counts (ablation)DVPG · MuJoCo continuous-control benchmarkssample efficiency and episode rewardImproved
Use 2 rollout steps rather than other tested rollout-step counts (ablation)DVPG · HumanoidStandup-v2, Hopper-v2, Ant-v2episode rewardImproved

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