QUYNH NGUYEN NGOC
Researcher/Ph.D. Student,
Faculty of Mathematics and Computer Science,
Saarland University
Address:
Building E 1 1, Room 229
Universität des Saarlandes
Postfach 15 11 50
D - 66041 Saarbrücken
Germany
phone:
quynh(at)cs(dot)uni-saarland(dot)de
ABOUT ME
I got my MS degree in Computer Science in 2015 from Saarland University where I have been supported by the International Max-Planck School for Computer Science (IMPRS). Before that, I received a BS in Computer Science from University of Engineering and Technology (a memeber of Vietnam National University Hanoi). My general research interests are in optimization, machine learning and computer vision.
My current focus is on non-convex optimization, in particular deep neural networks.
PUBLICATIONS
- Q. Nguyen and M. Hein
The Loss Surface and Expressivity of Deep Convolutional Neural Networks
preprint, available at arXiv - Q. Nguyen and M. Hein
The Loss Surface of Deep and Wide Neural Networks
ICML 2017. PDF
(Long version including proofs: PDF ) - A. Gautier, Q. Nguyen and M. Hein
Globally Optimal Training of Generalized Polynomial Neural Networks with Nonlinear Spectral Methods
NIPS 2016. PDF
Supplement: PDF
(code will be available soon)
- Q. Nguyen, A. Gautier and M. Hein
Nonlinear Spectral Methods for Nonconvex Optimization with Global Optimality
NIPS'16 Workshop On Optimization for Machine Learning. PDF
- Y. Xian, Z. Akata, G. Sharma, Q. Nguyen, M. Hein and B. Schiele
Latent Embeddings for Zero-shot Classification
Spotlight at CVPR 2016. arXiv
(project page) - Q. Nguyen, F. Tudisco, A. Gautier and M. Hein
An Efficient Multilinear Optimization Framework for Hypergraph Matching
accepted at IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 2016, preprint available at arXiv - Q. Nguyen, A. Gautier and M. Hein
A Flexible Tensor Block Coordinate Ascent Scheme for Hypergraph Matching
CVPR 2015 (oral presentation). PDF
Supplement: PDF
Extended Abstract: PDF
Poster: PDF
Slide: PDF
(project page)