报告题目：
Conditional graph entropy as an alternating minimization problem
报告时间：
2023年7月18日 15:30-16:30
报告地点：
腾讯会议室 ID：813 9361 1619
报告人：
Viktor Harangi
报告人简介：
Viktor Harangi received his Ph.D. in mathematics from ELTE Eötvös University in 2011. After a postdoctoral fellowship at University of Toronto, he returned to Budapest, Hungary, and is a researcher at the Rényi Institute.
报告简介：
Graph entropy as an alternating optimization problemKörner introduced the notion of graph entropy in 1973 as the minimal code rate of a natural coding problem where not all pairs of letters can be distinguished in the alphabet. He provided a simple, non-asymptotic formula. Later it turned out that it can also be expressed as the solution of a minimization problem over the so-called vertex-packing polytope. In this talk we show that these two formulas actually stem from the same alternating optimization problem. We discuss the resulting iterative algorithm converging to graph entropy. Joint work with Xueyan Niu and Bo Bai.
