UtilitySoftActorCritic

This repository includes the code to replicate the USAC algorithm from paper:

Improving Actor-Critic Training with Steerable Action-Value Approximation Errors
Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir, 2025
European Conference on Artificial Intelligence
ArXiv

USAC

To train the usac on cartpole swingup environment run the following command:

python main.py

Cite as

If you use USAC, please cite: