Ruslan Salakhutdinov

TL;DR: A new approximate inference algorithm for Deep Boltzmann Machines (DBM’s), a generative model with many layers of hidden variables, that learns a separate “recognition” model that is used to quickly initialize, in a single bottom-up pass, the values of the latent variables in all hidden layers.

TL;DR: The aim of the thesis is to demonstrate that deep generative models that contain many layers of latent variables and millions of parameters can be learned efficiently, and that the learned high-level feature representations can be successfully applied in a wide spectrum of application domains, including visual object recognition, information retrieval, and classification and regression tasks.

TL;DR: In this paper, the authors study the problem of designing models for machine learning tasks defined on sets and provide a family of functions to which any permutation invariant objective function must belong.

TL;DR: In this article, a constraint-based approach to visualizing high dimensional data is proposed to analyze the effect of parameter choices on data transformations and show that deep learning methods are able to learn physiologically important representations and detect latent relations in neuroimaging data.

TL;DR: Comprehensive experiments on both aligned and non-aligned multimodal time-series show that the MulT model outperforms state-of-the-art methods by a large margin, and empirical analysis suggests that correlated crossmodal signals are able to be captured by the proposed cross modal attention mechanism in MulT.