悉尼科技大学徐亦达副教授学术报告

来源:信息科学与技术学院  作者:朱焱  日期:2019-12-19  点击数:589

报告题目:Recent Research Involving Softmax Function

报告人: 徐亦达 (Richard Yida Xu),University of Technology, Sydney (UTS), Australia

报告时间:2019年12月20日(星期五) 上午9-11点

报告地点:西南交通大学犀浦校区4号教学楼X4131

主持人: 朱焱教授

 

报告内容摘要:

The softmax function has been used as the go-to method in classification, and has been used even more popular as the last output layer of a deep learning model. In this tutorial style talk, I will present several recent research works on machine learning invoking softmax function:

1. Probability re-paramterization using Gumb-max trick

2. Noise contrastive estimation for NLP

3. Open set classification.

报告人简介:

Richard Yi Da Xu is an associate professor in machine learning at school of electrical and data engineering of University of Technology Sydney. He leads a team of 30 people, include postdoc, PhD students and data engineers. His main research is in probabilistic machine learning, deep learning and compute vision. He published at many international conference and journals including AAAI, IJCAI and AISTATS, IEEE-(TNNLS,TIP,TKDE, and T-cybernetics). Since 2009, he published 1500+ slides of PhD training material in machine learning as well as many online ML videos. (richardxu.com). His team has collaborated with many Australian and overseas industries, including banks, e-commence, government, utilities, and legal sectors. He established a deep learning Sydney meetup, currently has 4300+ members. He is the head of Australian delegation to attend ISO JTC1 Artificial intelligence’s first plenary.  


欢迎广大师生参听!