List your company

Classic Discrete Control on CartPole v1

495Mean Episodic ReturnFP16-PPO
5.16132.33259.5386.67Apr 27, 2026May 2, 2026May 7, 2026May 12, 2026May 17, 2026May 22, 2026May 27, 2026

Evaluation results

MethodLinks
FP16-PPOPrecision=FP162026.04495—
INT8-PPOPrecision=INT82026.04492—
MLP-PPOArchitecture=MLP2026.04489—
INT4-PPOPrecision=INT42026.04487—
BitRL-1bitPrecision=1-bit2026.0447696.2
FedQHDEncoder Setting=Homo2026.05466.3—
Oracle QHD†Encoder Setting=Hetero2026.05458.5—
Oracle QHD†Encoder Setting=Homo2026.05373.2—
FedQHDEncoder Setting=Hetero2026.05351.1—
Trunc. FedAvg-QHDEncoder Setting=Homo2026.05230.8—
Oracle DQN†Encoder Setting=Hetero2026.05204.4—
Distill. FedDQNEncoder Setting=Homo2026.05169.9—
Distill. FedDQNEncoder Setting=Hetero2026.05149.4—
Oracle DQN†Encoder Setting=Homo2026.05139.1—
FedAvg-DQNEncoder Setting=Homo2026.05120.8—
Trunc. FedAvg-QHDEncoder Setting=Hetero2026.05112.6—
Indep. QHDEncoder Setting=Homo2026.0526.1—
Indep. QHDEncoder Setting=Hetero2026.0524—