Research map / Reinforcement learning / Policy gradient and actor-critic
VIREL: A Variational Inference Framework for Reinforcement Learning
Matthew Fellows, Anuj Mahajan, Tim G. J. Rudner, Shimon Whiteson
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use VIREL or beta instead of SAC | SAC · Humanoid-v2 and Ant-v2 | episodic return | Improved |
| Use VIREL or beta instead of DDPG | DDPG · Humanoid-v2 and Ant-v2 | episodic return | Improved |
| Use VIREL or beta instead of SAC and DDPG | SAC and DDPG · Simple low-dimensional MuJoCo tasks | episodic return | No clear effect |
| Use VIREL or beta instead of SAC | SAC · Half-Cheetah | episodic return | No clear effect |
| Use VIREL1 or VIREL2 instead of SAC | SAC with two soft critics · Humanoid-gym and Humanoid-rllab | episodic return | Improved |
| Use an estimated adaptive entropy scale instead of a constant entropy scale (ablation) | VIREL actor update · High-dimensional MuJoCo tasks | episodic return | No clear effect |
| Use two hard Q-functions instead of the standard VIREL critic setup (ablation) | VIREL1 · Half-Cheetah and humanoid tasks | episodic return | Mixed |
| Use one hard and one soft Q-function (ablation) | VIREL2 · Half-Cheetah and humanoid tasks | episodic return | Mixed |
| Remove the scalar lambda that scales the estimated residual error (ablation) | beta · MuJoCo continuous-control tasks | performance stability | Hurt |
Closest papers by meaning
- Probabilistic Inference in Reinforcement Learning Done Right (NeurIPS 2023 · similarity 0.91)
- Making Sense of Reinforcement Learning and Probabilistic Inference (ICLR 2020 · similarity 0.91)
- ReLU to the Rescue: Improve Your On-Policy Actor-Critic with Positive Advantages (ICML 2024 · similarity 0.88)
- Exploration from a Primal-Dual Lens: Value-Incentivized Actor-Critic Methods for Sample-Efficient Online RL (NeurIPS 2025 · similarity 0.88)
- Variational Delayed Policy Optimization (NeurIPS 2024 · similarity 0.87)
- Actor-Critics Can Achieve Optimal Sample Efficiency (ICML 2025 · similarity 0.87)
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.
