Posted on

cs535 special permission or prerequisite override requests for Fall 2018

If you are interested in registering for my Fall 2018  cs535 Pattern Recognition and do not have the prerequisites or for some other reason need a Special Permission (SPN) or Prerequisite Override, you will need to fill out a request here:

cs535 Fall 2018 SPN & Prereq Request Form

Note that you will need a Google account in order to sing in and see the form.

Please do not email me with individual requests.  I will be issuing SPNs and Prerequisite Overrides no earlier than 2 weeks before the start of the Fall 2018 semester.

Posted on

ICCV 2017 – Wrap Up

Last week was the time for ICCV 2017 in Venice, Italy.  Setting aside my personal indifference to Venice (overcrowded and overpriced, uninspiring food),  ICCV was an interesting meeting.  It took place on Lido, at the same venue as the Venice Film Festival.  It was certainly a more appealing location to me personally, but a bit of a way from San Marco and the tourist Venice.

The conference itself was, well, a mixed bag.  Plenty of deep block shuffling works.  About 3000 attendees is the number I heard.

We presented our work “PUnDA: Probabilistic Unsupervised Domain Adaptation for Knowledge Transfer Across Visual Categories” [1]. It shows that it is not always necessary to have end-to-end learning if the transfer model is properly constructed.   As the luck has it, I had to reshuffle the schedule so the poster was both on Wednesday and Thursday (thanks to Robert Walecki, who help with the poster.)  You can find a copy of the poster below.

[1] B. Gholami, O. Rudovic, and V. Pavlovic, “PUnDA: Probabilistic Unsupervised Domain Adaptation,” in Proc. IEEE International Conference Computer Vision, 2017.
[Bibtex]
@inproceedings{behnam17iccv,
added-at = {2017-07-18T17:26:03.000+0200},
author = {Gholami, Behnam and Rudovic, Ognjen and Pavlovic, Vladimir},
biburl = {https://www.bibsonomy.org/bibtex/29fd2aed9b960e71bf2472b5d70079060/vpavlovic},
booktitle = {Proc. IEEE International Conference Computer Vision},
interhash = {4a8c0ac9cfbd7b02e2d86ef7b5a289a0},
intrahash = {9fd2aed9b960e71bf2472b5d70079060},
keywords = {domain_adaptation myown unsupervised},
owner = {vladimir},
timestamp = {2017-07-18T18:21:19.000+0200},
title = {PUnDA: Probabilistic Unsupervised Domain Adaptation},
year = 2017
}
Posted on

cs536 – Machine Learning, Spring 2018 – Registration Requests

If you are interested in registering for my Spring 2018 Machine Learning course (01:198:536) and do not have the prerequisites or for some other reason need a Special Permission (SPN) or Prerequisite Override, you will need to fill out a request here:

cs535 Spring 2018 SPN & Prereq Request Form

Note that you will need a Google account in order to sing in and see the form.

Please do not email me with individual requests.  I will be issuing SPNs and Prerequisite Overrides no earlier than 2 weeks before the start of the Spring 2018 semester.

Posted on

“Unsupervised Domain Adaptation with Copula Models” presented at MLSP’17

Our paper “Unsupervised Domain Adaptation with Copula Models”[1] was presented this week at the IEEE Int’l Workshop on Machine Learning for Signal Processing in Tokyo, Japan.

It was an exciting meeting; unlike the now mega-conferences of CVPR and NIPS kind, MLSP is still refreshingly small.  There were several outstanding tutorials and keynotes by Kenji Fukumizu, Shun-ichi Amari, and Yee Whye Teh, with limited emphasis on “deep.”  Well organized!

Then, there was Tokyo itself.  Always a pleasure to visit.

References

[1] C. D. Tran, O. Rudovic, and V. Pavlovic, “Unsupervised domain adaptation with copula models,” in IEEE Int’l Conf. Machine Learning for Signal Processing (MLSP), 2017.
[Bibtex]
@InProceedings{tran17mlsp,
author = {Cuong D. Tran and Ognjen Rudovic and Vladimir Pavlovic},
title = {Unsupervised domain adaptation with copula models},
booktitle = {IEEE Int’l Conf. Machine Learning for Signal Processing (MLSP)},
year = {2017},
note = {33\% contribution.},
date-added = {2017-01-17 15:43:46 +0000},
date-modified = {2017-01-17 15:45:08 +0000},
keywords = {domain adaptation, mlsp17},
}
Posted on

Chapter on “Machine Learning Methods for Social Signal Procesing” in Social Signal Processing book

Social Signal Processing book by Cambridge University Press was finally published this July.   Read our chapter on “Machine Learning Methods for Social Signal Processing”[1] on p. 234 of this collection of outstanding articles.

References

[1] O. Rudovic, M. Nicolaou, and V. Pavlovic, “Social Signal Processing,” , A. Vinciarelli, J. Burgoon, N. Magnenat-Thalmann, and M. Pantic, Eds., Cambridge University Press, 2017.
[Bibtex]
@inbook{rudovic14mlssp,
Author = {Ognjen Rudovic and Mihalis Nicolaou and Vladimir Pavlovic},
Chapter = {Machine Learning Methods for Social Signal Processing},
Editor = {Alessandro Vinciarelli and Judee Burgoon and Nadia Magnenat-Thalmann and Maja Pantic},
Note = {33\% contribution},
Publisher = {Cambridge University Press},
Title = {Social Signal Processing},
Year = {2017}}
Posted on

Welcome to new lab members

This fall we are welcoming two new students to our group:  Mihee Lee and Yuting Wang.   Mihee joins us from R&D at Samsung, where she worked after completing her BS in Math Ewha Woman’s University, Korea.  Yuting completed her MS at the Karlsruhe Institute of Technology in Germany and was also a visiting student at CMU.

Please join me in welcoming Mihee and  Yuting to Rutgers and Seqam Lab.

Posted on

Special Permissions / Overrides for Fall 17 Courses

Many students have contacted me asking about Special Permissions or  Prerequisite Overrides for 206 & 535 courses for Fall 2017.  As of now, Thursday, August 31, the status is as follows:  I am told that our administrative staff is still processing requests.  I have yet to receive any lists of students who submitted their requests.

Once I have more information, I will post it here.

Update (9/5/2017):  This afternoon I finally received access to SPN request lists.  I will be processing those requests in the next few days.

Posted on

CVPR 2017 – Wrap Up

It was quite exciting to attend the largest CVPR ever – almost 5000 attendees.  Having it in a beautiful location made it even more appealing.

Thanks to my students and colleagues who made the work we presented at CVPR possible.

Joint work with Imperial College and MIT on using copula models for joint facial AU estimation.
Joint NTU Singapore – Rutgers work on generative models for robust 3D face pose estimation
Break time at Waikiki beach
Hai presenting his work at the 1st Int’l Workshop on Deep Affective Learning and Context Modeling