Kuhn Poker
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Exploitability computation | 256-Kuhn poker (full) | Exploitability0.007 | 30 | |
| Payoff prediction (Task A - Fixed) | Kuhn Poker | MSE0.0208 | 10 | |
| Multi-Agent Strategic Reasoning | Kuhn Poker In-domain | Success Rate65.6 | 8 | |
| Payoff prediction (Task B) | Kuhn Poker | MSE0.035 | 6 | |
| Game Playing | Kuhn Poker | Raw Reward0.98 | 6 | |
| Solving Zero-Sum Partially Observable Stochastic Games (zs-POSGs) | Kuhn Poker | Time0.01 | 6 | |
| Agent identification | Kuhn Poker 500 eval agents | Top-1 Accuracy47.9 | 4 | |
| Agent identification | Kuhn Poker (eval) | Top-1 Accuracy47.9 | 4 | |
| Game Playing | Kuhn Poker FF Game (test) | Exploitability0.009 | 4 | |
| Game Playing | Kuhn Poker Standard (test) | Exploitability0.013 | 4 | |
| CFR solver performance | Kuhn poker | Average Iteration Time1.04 | 4 | |
| Best-response payoff prediction | Kuhn Poker | Payoff (Best-Response)0.472 | 3 | |
| Payoff prediction (Task C) | Kuhn Poker | MSE0.036 | 3 | |
| Payoff prediction (Task B - Pair) | Kuhn Poker | MSE0.035 | 3 | |
| Strategic game playing | Kuhn Poker (train) | Win Rate44.49 | 2 |
Showing 15 of 15 rows