Research map / Reinforcement learning / Visual and representation RL
VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning
Luisa Zintgraf, Kyriacos Shiarlis, Maximilian Igl, Sebastian Schulze, Yarin Gal, Katja Hofmann, 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 a stochastic task embedding and infer its posterior from experience | variBAD · Gridworld | average return | Improved |
| Use a stochastic task embedding and infer its posterior from experience | variBAD · MuJoCo continuous control tasks | online return | Improved |
| Condition action selection on the task posterior | variBAD · Gridworld | exploration efficiency | Improved |
| Decode future as well as past transitions and rewards | variBAD · Meta-RL tasks | inference about unseen states | Improved |
| Remove the decoder and VAE objective, leaving a recurrent policy (ablation) | variBAD · Meta-RL | task representation | Hurt |
| Train with a reward decoder only rather than also a dynamics decoder (ablation) | variBAD · MuJoCo continuous control tasks | performance | Improved |
| Do not backpropagate the RL loss through the encoder (ablation) | variBAD · Meta-RL training | training speed | Improved |
| Do not backpropagate the RL loss through the encoder (ablation) | variBAD · Meta-RL training | policy learning | No clear effect |
| Replace the learned stochastic latent belief and decoder with an RL2-style recurrent policy (ablation) | variBAD · Gridworld | matching Bayes-optimal solutions across seeds | Mixed |
Closest papers by meaning
- Bayesian Exploration Networks (ICML 2024 · similarity 0.85)
- Learning Task Belief Similarity with Latent Dynamics for Meta-Reinforcement Learning (ICLR 2025 · similarity 0.85)
- Provable and Practical: Efficient Exploration in Reinforcement Learning via Langevin Monte Carlo (ICLR 2024 · similarity 0.85)
- MAMBA: an Effective World Model Approach for Meta-Reinforcement Learning (ICLR 2024 · similarity 0.85)
- Exploration from a Primal-Dual Lens: Value-Incentivized Actor-Critic Methods for Sample-Efficient Online RL (NeurIPS 2025 · similarity 0.85)
- DynaMITE-RL: A Dynamic Model for Improved Temporal Meta-Reinforcement Learning (NeurIPS 2024 · similarity 0.84)
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.
