报告题目：Marrying Causality and Association with Visual Learning
时间：12月05日周一上午10:00-11:30
地点：计算所446会议室
报告摘要：Compared with machines, people are far better learners as they are capable of learning models from very limited samples of a new category and make accurate predictions and reasoning accordingly. The strong learning ability of human are attributed to the key elements of causality and association in the human learning process. To learn from human learning process is a promising way to develop new effective learning methods. In this talk, several basic problems in human-like visual learning will be discussed, including the bottle-neck of modern visual learning methods, the notion of causality and association in visual learning, and some of our preliminary trails and experiences on marrying causality and association with visual learning.
报告人简介：
崔鹏，清华大学计算机系长聘副教授，博士生导师，2010年于清华大学获得博士学位。研究领域包括大数据环境下的因果推理与稳定预测、网络表征学习、社会动力学建模，及其在金融科技、智慧医疗及社交网络等场景中的应用。已在数据挖掘及多媒体领域顶级国际期刊和会议上发表论文百余篇，并先后获得7项国际会议及期刊最佳论文奖，包括中国入选数据挖掘领域顶级国际会议KDD最佳论文专刊的首篇论文。目前担任IEEE TKDE、IEEE TBD、ACM TOMM等三个Transactions期刊的编委。获得中国计算机学会青年科学家奖，国际计算机协会（ACM）中国新星奖，并入选中国科协首届青年人才托举计划。曾先后获得教育部自然科学一等奖、北京市科技进步一等奖、中国电子学会自然科学一等奖。
附件：