SergioTermann

Reinforcement Learning douyin:9197123773 wechat:［手机号已隐藏］ bilibili:3493091679930873

代表作品 / 项目

dogfightEnv⭐ 101Dogfight RL environment based on Harfang3D dogfight sandbox
awesome-post-training-RL⭐ 51A curated list of papers on reinforcement learning post-training for LLMs
windrise⭐ 20
procgen-competition-2020⭐ 1NeurIPS 2020 Procgen Competition archive: Ray/RLlib modifications, data augmentation, per-round submissions and team retrospective (warm-up 10th, round-1 7th, round-2 top-16)
QIARL⭐ 1QIARL: representation learning with Q-pi irrelevance state abstraction for RL (PPO + auxiliary Q-value state-aggregation loss), evaluated on OpenAI Procgen

条目信息来自其公开主页与公开发布内容。按本站规范，页面不展示任何联系方式。需要更正或删除？通过收录与更正通道提交，24 小时内处理。