notesum.ai

Published at November 20

cs.RO

cs.AI

Released Date: November 20, 2024

Authors: Feng Gao, Chao Yu, Yu Wang, Yi Wu

Hyper Parameters for Training
Batch Size	$1024\times 128$
Mini-batch Size	1024
Number of Epochs	5
Clip Range	0.2
Entropy Coefficient	0.01
Discount Factor	0.99
GAE Discount Factor	0.95
Desired KL-Divergence	0.01
Learning Rate	$1\times 10^{-3}$
Adam Epsilon	$1\times 10^{-8}$
Gradient Clipping	10