In the Spring of 2016, I will be moving to the Machine Learning Department at Carnegie Mellon University. I am looking for strong PhD students, please apply to CMU if you are interested in working with me.
I am an assistant professor of Computer Science and Statistics at the University of Toronto. I work in the field of statistical machine learning (See my CV.) I received my PhD in computer science from the University of Toronto in 2009. After spending two post-doctoral years at MIT, I joined the University of Toronto in 2011.
My research interests include
Deep Learning,
Probabilistic Graphical Models, and
Large-scale Optimization.
Prospective students: Please
read this to ensure that I read your email.
Recent Research Highlights:
- See our recent Deep Learning Tutorial in Montreal:
Part 1:[Slides (pdf)], [Video]
Part 2:[Slides (pdf)], [Video] - See our recent Deep Learning Tutorial at KDD 2014: [Video], [ Slides].
- Check out our new website with demos and software.
- I was helping to run Thematic Program on Statistical Inference, Learning, and Big Data at the Fields Institute.
- I was teaching an advanced Machine Learning course at the Fields Institute. Videos of my lectures will be available online. Also, check out Live Streaming of my course.
Recent Papers:
-
Gated-Attention Readers for Text Comprehension
Bhuwan Dhingra, Hanxiao Liu, William W. Cohen, Ruslan Salakhutdinov
arXiv [arXiv]. -
Deep Neural Networks with Massive Learned Knowledge
Zhiting Hu, Zichao Yang, Ruslan Salakhutdinov, and Eric Xing
To appear in the Conference on Empirical Methods in Natural Language Processing (EMNLP'16). -
Iterative Refinement of Approximate Posterior for Training Directed Belief Networks
Devon Hjelm, Kyunghyun Cho, Junyoung Chung, Russ Salakhutdinov, Vince Calhoun, Nebojsa Jojic
To appear in NIPS 2016, arXiv [arXiv]. -
Path-Normalized Optimization of Recurrent Neural Networks with ReLU Activations
Behnam Neyshabur, Yuhuai Wu, Ruslan Salakhutdinov, Nathan Srebro
To appear in NIPS 2016, arXiv [arXiv]. -
Stochastic Variational Deep Kernel Learning
Andrew Gordon Wilson, Zhiting Hu, Eroc Xing, Ruslan Salakhutdinov
To appear in NIPS 2016. -
On Multiplicative Integration with Recurrent Neural Networks
Yuhuai Wu, Saizheng Zhang, Ying Zhang, Yoshua Bengio, Ruslan Salakhutdinov
To appear in NIPS 2016, arXiv [arXiv]. -
Encode, Review, and Decode: Reviewer Module for Caption Generation
Zhilin Yang, Ye Yuan, Yuexin Wu, Ruslan Salakhutdinov, William W. Cohen
To appear in NIPS 2016, arXiv [arXiv]. -
Architectural Complexity Measures of Recurrent Neural Networks
Saizheng Zhang, Yuhuai Wu, Tong Che, Zhouhan Lin, Roland Memisevic, Ruslan Salakhutdinov, Yoshua Bengio
To appear in NIPS 2016, arXiv [arXiv]. -
Multi-Task Cross-Lingual Sequence Tagging from Scratch
Zhilin Yang, Ruslan Salakhutdinov, William Cohen
arXiv [arXiv]. -
Revisiting Semi-Supervised Learning with Graph Embeddings
Zhilin Yang, William Cohen, Ruslan Salakhutdinov
ICML 2016, arXiv [arXiv]. -
Importance Weighted Autoencoders
Yuri Burda, Roger Grosse, Ruslan Salakhutdinov
ICLR, 2016, [arXiv]. Code is available [here]. -
Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning
Emilio Parisotto, Jimmy Lei Ba, Ruslan Salakhutdinov
ICLR, 2016, [arXiv]. -
Generating Images from Captions with Attention
Elman Mansimov, Emilio Parisotto, Jimmy Lei Ba, Ruslan Salakhutdinov
ICLR, 2016, oral [arXiv]. [Generated Samples].- See also
Bloomberg news,
Motherboard.
- See also
Bloomberg news,
Motherboard.
-
Data-Dependent Path Normalization in Neural Networks
Behnam Neyshabur, Ryota Tomioka, Ruslan Salakhutdinov, Nathan Srebro
ICLR, 2016, [arXiv]. -
Action Recognition using Visual Attention
Shikhar Sharma, Ryan Kiros, Ruslan Salakhutdinov
ICLR workshop, 2016 [arXiv]. [Code]. [Project Website]. -
Deep Kernel Learning
Andrew Gordon Wilson, Zhiting Hu, Ruslan Salakhutdinov, Eric Xing
AI and Statistics, 2016, [arXiv]. -
Human-level concept learning through probabilistic program induction
Brenden Lake, Ruslan Salakhutdinov, and Joshua Tenenbaum (2015),
Science, 350(6266), 1332-1338, [paper], [Supporting Info.], [visual Turing tests], [Omniglot data set], [Code].- See also New York Times, CBC, Reuters, CBS, MIT Tech Review, Toronto Star, UofT News, MIT News, Washington Post, CIFAR, Business Insider.
-
Learning Wake-Sleep Recurrent Attention Models
Lei Jimmy Ba, Roger Grosse, Ruslan Salakhutdinov, Brendan Frey
NIPS 2015. [arXiv]. -
Skip-Thought Vectors
Ryan Kiros, Yukun Zhu, Ruslan Salakhutdinov, Richard S. Zemel, Antonio Torralba, Raquel Urtasun, Sanja Fidler
NIPS 2015, [arXiv]. -
Path-SGD: Path-Normalized Optimization in Deep Neural Networks
Behnam Neyshabur, Ruslan Salakhutdinov, Nathan Srebro
NIPS 2015, [arXiv]. -
Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books
Yukun Zhu, Ryan Kiros, Richard Zemel, Ruslan Salakhutdinov, Raquel Urtasun, Antonio Torralba, Sanja Fidler
ICCV 2015, [arXiv], oral , [ project page ] -
Predicting Deep Zero-Shot Convolutional Neural Networks using Textual Descriptions
Jimmy Ba, Kevin Swersky, Sanja Fidler, Ruslan Salakhutdinov
ICCV 2015, [arXiv]. -
Learning Deep Generative Models
Ruslan Salakhutdinov
Annual Review of Statistics and Its Application, Vol. 2, pp. 361–385, 2015
[pdf]. -
Scaling Up Natural Gradient by Sparsely Factorizing the Inverse Fisher Matrix
Roger Grosse, Ruslan Salakhutdinov
ICML, 2015. [pdf]. -
Unsupervised Learning of Video Representations using LSTMs
Nitish Srivastava, Elman Mansimov, Ruslan Salakhutdinov
ICML, 2015, [arXiv], [Code] -
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron Courville, Ruslan Salakhutdinov, Richard Zemel, Yoshua Bengio
ICML, 2015, [arXiv], [project page],- See also
Scientific American,
CIFAR.
- See also
Scientific American,
CIFAR.
-
Siamese neural networks for one-shot image recognition.
Gregory Koch, Richard Zemel, Ruslan Salakhutdinov
ICML 2015 Deep Learning Workshop (2015), [pdf]. -
Exploiting Image-trained CNN Architectures for Unconstrained Video Classification
Shengxin Zha, Florian Luisier, Walter Andrews, Nitish Srivastava, Ruslan Salakhutdinov
BMVC 2015, [arXiv], 2015 -
segDeepM: Exploiting Segmentation and Context in Deep Neural Networks for Object Detection
Y. Zhu, R. Urtasun, R. Salakhutdinov and S.Fidler
In Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, June 2015,
[ arXiv ] -
Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
Ryan Kiros, Ruslan Salakhutdinov, Richard Zemel.
To appear in Transactions of the Association for Computational Linguistics (TACL), 2015.
[ arXiv], [ results], [ demo ].
An encoder-decoder architecture for ranking and generating image descriptions.
Previous version appeared in NIPS Deep Learning Workshop, 2014. -
Accurate and Conservative Estimates of MRF Log-likelihood using Reverse Annealing
Yuri Burda, Roger B. Grosse, and Ruslan Salakhutdinov,
AI and Statistics, 2015 [arXiv]