List your company

Classic Discrete Control on Acrobot v1

94Mean Episodic ReturnBitRL-1bit
81.5284.768891.24Apr 27, 2026

Evaluation results

MethodLinks
BitRL-1bitPrecision=1-bit2026.049487.1
INT4-PPOPrecision=INT42026.0489—
MLP-PPOArchitecture=MLP2026.0488—
INT8-PPOPrecision=INT82026.0485—
FP16-PPOPrecision=FP162026.0482—