Search For:

Displaying 1-5 out of 5 total
Unsupervised Learning of Image Transformations
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Roland Memisevic, Geoffrey Hinton
Issue Date:June 2007
pp. 1-8
We describe a probabilistic model for learning rich, distributed representations of image transformations. The basic model is defined as a gated conditional random field that is trained to predict transformations of its inputs using a factorial set of late...
Gradient-based learning of higher-order image features
Found in: Computer Vision, IEEE International Conference on
By Roland Memisevic
Issue Date:November 2011
pp. 1591-1598
Recent work on unsupervised feature learning has shown that learning on polynomial expansions of input patches, such as on pair-wise products of pixel intensities, can improve the performance of feature learners and extend their applicability to spatio-tem...
The Potential Energy of an Autoencoder
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Hanna Kamyshanska,Roland Memisevic
Issue Date:February 2015
pp. 1
Autoencoders are popular feature learning models, that are conceptually simple, easy to train and allow for efficient inference and training. Recent work has shown how certain autoencoders can be associated with an energy landscape, akin to negative log-pr...
Combining modality specific deep neural networks for emotion recognition in video
Found in: Proceedings of the 15th ACM on International conference on multimodal interaction (ICMI '13)
By Çaglar Gülçehre, Guillaume Desjardins, Jeremie Zumer, Mehdi Mirza, Nicolas Boulanger-Lewandowski, Sébastien Jean, Yann Dauphin, Aaron Courville, Abhishek Aggarwal, Arjun Sharma, Atousa Torabi, Christopher Pal, David Warde-Farley, Emmanuel Bengio, Jean-Philippe Raymond, Kishore Reddy Konda, Pascal Lamblin, Pascal Vincent, Pierre Froumenty, Pierre-Luc Carrier, Raul Chandias Ferrari, Razvan Pascanu, Roland Memisevic, Samira Ebrahimi Kahou, Xavier Bouthillier, Yoshua Bengio, Zhenzhou Wu
Issue Date:December 2013
pp. 543-550
In this paper we present the techniques used for the University of Montréal's team submissions to the 2013 Emotion Recognition in the Wild Challenge. The challenge is to classify the emotions expressed by the primary human subject in short video clips...
Kernel information embeddings
Found in: Proceedings of the 23rd international conference on Machine learning (ICML '06)
By Roland Memisevic
Issue Date:June 2006
pp. 633-640
We describe a family of embedding algorithms that are based on nonparametric estimates of mutual information (MI). Using Parzen window estimates of the distribution in the joint (input, embedding)-space, we derive a MI-based objective function for dimensio...