Reinforcement Learning on Acrobot (standard)
850Model ParametersAlgorithm 1 (GMM-QFs)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Algorithm 1 (GMM-QFs)K=50, Gaussian functions=502025.12 | 850 | — | — | |
| Algorithm 1 (GMM-QFs)K=500, Gaussian functions=5002025.12 | 8,500 | — | — | |
| DQNNetwork architecture=2 128-neuron layers2025.12 | 17,795 | — | — | |
| PPONetwork architecture=2 networks, 2 128-neuron layers2025.12 | 35,332 | — | — |