机器学习与统计学（MLSTAT）会议是由中国现场统计研究会机器学习分会主办的学术会议。会议旨在促进机器学习与统计领域的国内外学者进行学术交流，引领机器学习与统计共同交叉发展的学术文化，推动作为数据科学与人工智能的奠基性学科的进步，以此助力相关数字经济产业的发展。
第四届机器学习与统计学会议（MLSTAT2026）将于2026年7月15日-17日在西南财经大学（四川成都市）举行。会议将邀请20位左右青年学者就机器学习、人工智能、统计学和应用数学等相关领域的前沿进展做大会主题报告，并安排有Tutorial, 同时欢迎大家进行墙报展示。
时间：7月15日（周三）下午14：00-18：00
地点：柳林校区（具体地点后续通知）
课程摘要:
This short course provides a unified introduction to reinforcement learning (RL) and its central role in modern large language models (LLMs). The first part covers the foundations of RL. Special emphasis is placed on policy optimization methods such as PPO, which forms the basis of algorithms widely used in LLM training. The second part turns to LLMs, especially LLM alignment where RL has been particularly influential. We will introduce methods such as RLHF and DPO for aligning LLMs with human preferences. The course is designed for researchers, practitioners, and graduate students seeking both foundational knowledge and insights into current frontiers.
主讲人:
Chengchun Shi is an Associate Professor in the Department of Statistics at the London School of Economics and Political Science. His research interests lie in the interaction between statistics, RL, and LLMs. He has published widely in leading journals in statistics and highly-ranked computer science. Dr. Shi has taught an RL short course https://github.com/callmespring/RL-short-course at various universities, which has attracted around 80 stars on GitHub over the past few years. At LSE, he has taught multiple graduate-level courses on machine learning, deep learning and RL, and brings both theoretical expertise and practical experience to the course. In recognition of his contributions to teaching, he received the Excellence in Education Award from LSE for the 2021–2022 academic year.
会议注册信息:
为了确保会议顺利开展，本次会议将少量收取注册费，会务组将承担会议期间的用餐，其他费用敬请自理！
注册费用：学生代表 200元，其他代表500元。
报名截止时间：2026年6月30日。
会议信息将通过本公众号及会议网站及时更新，欢迎大家积极关注。也请已报名的同学和老师扫码补充tutorial信息，会议注册是通过会议网站注册。会议注册网址和注册二维码：https://ml-stat.github.io/MLSTAT2026/register/
联系方式:
邮箱：mlstat2026@126.com
电话：028-87092330
