Zero-Sum Games
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Nash Equilibrium Approximation | Zero-sum games Dimension 7 | Iterations (10% Rel. Error)133 | 9 | |
| Nash equilibrium computation | Zero-sum games 5x5 | Success Rate @ 0.10100 | 9 | |
| Nash equilibrium computation | Zero-sum games 7x7 | Success Rate (s@0.10)100 | 8 | |
| Nash equilibrium computation | Zero-sum games 4x4 | s@0.101 | 8 | |
| Nash equilibrium computation | Zero-sum games 3x3 | s@0.10 Success Rate100 | 8 | |
| Nash equilibrium computation | Zero-sum games 2x2 | s@0.10 Score100 | 8 | |
| Nash equilibrium computation | Zero-sum games 6x6 | Success Rate @ 0.10100 | 7 | |
| Multi-agent Reinforcement Learning | Zero-sum games | Average KL Divergence0.001 | 5 | |
| Approximating Nash Equilibria | Zero-sum games small learning rates Dimension 5, 10, 15, 20 | Average Relative Error (1 min)0.0001 | 4 | |
| Finding epsilon-Approximate Nash Equilibria | Zero-Sum Games | Regret1 | 3 |
Showing 10 of 10 rows