[Adapt] Last seminar this Wednesday
luo.kangqi at qq.com
Wed Jun 7 02:03:42 CST 2017
This time, I'll mainly introduce some GAN's variants, namely CGAN and IRGAN.
Time: 4:30 pm, Jun. 7th
Location: SEIEE 3-404
GAN (Generative Adversarial Nets) is a hot research topic in recent machine learning fields, since the underlying intuition of reinforcement learning has brought lot of success, especially in CV area.
IRGAN is a new paper published in SIGIR 2017, and one of our ADAPTer, Yu Gong, made much contributions to the paper! IRGAN proposed an adversarial learning framework into the traditional IR task, and I believe this framework will be helpful in many supervised learning scenarios.
If you are not familiar with GAN, please check out Haijun's seminar slides in advance.
Another GAN's tutorial (By Yanran Li, PolyU) can be downloaded here, which is VERY VERY informative.
See you then, enjoy our last seminar!
Kangqi Luo, PhD Candidate
ADAPT Lab, Department of Computer Science
SEIEE 3-341, Shanghai Jiao Tong University,
No. 800 Dongchuan Road, Shanghai, China
------------------ Original ------------------
From: "Kenny Zhu";<kzhu at cs.sjtu.edu.cn>;
Date: Mon, Jun 5, 2017 10:31 PM
To: "adapt"<adapt at cs.sjtu.edu.cn>;
Subject: [Adapt] Last seminar this Wednesday
This week marks the 16th week of the semester and we will have our last group seminar. Kangqi will present a talk on GAN this time. As a tradition, we will have a photo session at the end. I hope everybody can attend this one so that our half-yearly photo includes your smile!
See you on Wednesday,
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Adapt