Train Board Snapshot
Top Ranked Research Bets
This host now serves baked-in Train Board data directly, without loading policy-dashboard APIs.
| Task | Blended | Impact | Execution |
|---|---|---|---|
| Robust Autonomy Emerges from Self-Play | 77.3% | 87.5% | 58.3% |
| Muesli: Combining Improvements in Policy Optimization | 73.3% | 80.0% | 60.8% |
| TD-MPC2: Scalable, Robust World Models for Continuous Control | 72.9% | 82.5% | 55.0% |
| AlphaStar: Grandmaster level in StarCraft II using multi-agent reinforcement learning | 72.8% | 85.0% | 50.0% |
| How To Scale Your (Transformer) Model | 72.3% | 81.7% | 55.0% |
| Podracer architectures for scalable Reinforcement Learning | 71.3% | 78.3% | 58.3% |
| Multi-Agent Reinforcement Learning: Foundations and Modern Approaches | 71.3% | 80.0% | 55.0% |
| Human-Timescale Adaptation in an Open-Ended Task Space (ADA) | 71.2% | 81.7% | 51.7% |