DiCo
Cross-source consensus on DiCo from 1 sources and 5 claims.
1 sources · 5 claims
Benefits
Preparation
Comparisons
Evidence quality
Highlighted claims
- In DiCo experiments, empty Bernoulli draws fall back to the full pair set so the control ratio remains defined. — Graph-SND: Sparse Aggregation for Behavioral Diversity in Multi-Agent Reinforcement Learning
- At n=50, Bernoulli-0.1 Graph-SND used 122.5 expected sampled edges compared with 1,225 full pairs. — Graph-SND: Sparse Aggregation for Behavioral Diversity in Multi-Agent Reinforcement Learning
- Replacing full SND with sparse Graph-SND preserved closed-loop diversity control in VMAS Dispersion experiments. — Graph-SND: Sparse Aggregation for Behavioral Diversity in Multi-Agent Reinforcement Learning
- The DiCo head-to-head comparison reported metric speedups of about 9.24x to 9.63x with a small paired reward difference. — Graph-SND: Sparse Aggregation for Behavioral Diversity in Multi-Agent Reinforcement Learning
- At n=10, several sparse Graph-SND variants tracked the desired SND level and matched the full-SND controller's reward within seed variation. — Graph-SND: Sparse Aggregation for Behavioral Diversity in Multi-Agent Reinforcement Learning