Learn More
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 extracted from feature length movies. This involves the analysis of video clips of acted scenes(More)
Theano is a Python library that allows to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. Since its introduction, it has been one of the most used CPU and GPU mathematical compilers - especially in the machine learning community - and has shown steady performance improvements. Theano is being actively(More)
The ICML 2013 Workshop on Challenges in Representation Learning(1) focused on three challenges: the black box learning challenge, the facial expression recognition challenge, and the multimodal learning challenge. We describe the datasets created for these challenges and summarize the results of the competitions. We provide suggestions for organizers of(More)
ACKNOWLEDGEMENTS This study was conducted by the Working Group on Library Instruction of the Subcommittee on Libraries of the Conference of Rectors and Principals of Quebec Universities (CREPUQ), in cooperation with several other parties. We would like to thank the library Directors and Registrars of the participating universities for their invaluable(More)
The quasiharmonic approximation ͑QHA͒, in its simplest form also called the statically constrained ͑SC͒ QHA, has been shown to be a straightforward method to compute thermoelastic properties of crystals. Recently we showed that for noncubic solids SC-QHA calculations develop deviatoric thermal stresses at high temperatures. Relaxation of these stresses(More)
We apply the spike-and-slab Restricted Boltzmann Machine (ssRBM) to texture modeling. The ssRBM with tiled-convolution weight sharing (TssRBM) achieves or surpasses the state-of-the-art on texture synthesis and inpainting by parametric models. We also develop a novel RBM model with a spike-and-slab visible layer and binary variables in the hidden layer.(More)
—High Performance Computing (HPC) Best Practice offers opportunities to implement lessons learned in areas such as computational chemistry and physics in genomics workflows, specifically Next-Generation Sequencing (NGS) workflows. In this study we will briefly describe how distributed-memory parallelism can be an important enhancement to the performance and(More)
Learning tasks such as those involving genomic data often poses a serious challenge: the number of input features can be orders of magnitude larger than the number of training examples, making it difficult to avoid overfitting, even when using the known regularization techniques. We focus here on tasks in which the input is a description of the genetic(More)
  • 1