AAAI 2026technical0 citations
PT-DCFR: Accelerating and Improving Deep CFR Using Population Based Training (Student Abstract)
Dingzhong Cai, Huale Li, Hang Xiao, Shuhan Qi, Jiajia Zhang
Abstract
Deep CFR enables end-to-end approximation of Nash equilibria in imperfect-information games(IIGs) but is sensitive to hyperparameters, making manual tuning inefficient. To address this, we propose PT-DCFR, which integrates Population-Based Training(PBT) with Deep CFR to dynamically optimize hyperparameters during training. Building upon this, we further introduce P2T-DCFR, which decouples parameter selection from model performance.
BibTeX
@inproceedings{aaai2026_ptdcfraccelerati,
title = {PT-DCFR: Accelerating and Improving Deep CFR Using Population Based Training (Student Abstract)},
author = {Dingzhong Cai and Huale Li and Hang Xiao and Shuhan Qi and Jiajia Zhang},
booktitle = {AAAI 2026},
year = {2026}
}