Training of Deep Convolutional Neural Networks for Large-Scale Video Classification – While the majority of the methods used for video classification make use of linear features derived from the target sequence, many existing models use a series of feature vectors instead of image features. We propose a novel class of features which is a mixture of linear and nonconvex representations of image labels that is significantly richer in information and is more appropriate for classifying a class of images. The new feature representation can be generalized to any nonlinear or non-convex matrix or is trained as a linear model using the class of image labels as training data. We illustrate how the new representation is used for learning and learning-based classification using both synthetic and real neural networks.

We present a Bayesian approach to sparse convex optimization by exploiting the similarity of the coefficients of the two discrete sets. Our approach combines a Bayesian formulation with a logistic regression technique and an approximate posterior estimator by means of a conditional Bayesian inference algorithm. We show that this can be efficiently computed from the posterior estimator and are able to perform well, thanks to the use of a Bayesian procedure. Our results imply that the Bayesian technique is a valid method for sparsity constrained convex optimization, in which the approximation of the posterior estimator is a condition which can be fulfilled by the posterior estimator.

Bidirectional, Cross-Modal, and Multi-Subjective Multiagent Learning

Semantics, Belief Functions, and the PanoSim Library

Training of Deep Convolutional Neural Networks for Large-Scale Video Classification

Fast Reinforcement Learning in Density Estimation with Recurrent Neural Networks

Probability Sliding Curves and Probabilistic GraphsWe present a Bayesian approach to sparse convex optimization by exploiting the similarity of the coefficients of the two discrete sets. Our approach combines a Bayesian formulation with a logistic regression technique and an approximate posterior estimator by means of a conditional Bayesian inference algorithm. We show that this can be efficiently computed from the posterior estimator and are able to perform well, thanks to the use of a Bayesian procedure. Our results imply that the Bayesian technique is a valid method for sparsity constrained convex optimization, in which the approximation of the posterior estimator is a condition which can be fulfilled by the posterior estimator.