Connections between Retinex, neural models and variational methods

Variational Approach to Retinex Theory

Graphical Models and Variational Methods

Variational Methods for Graphical Models

∗a summary of [JGJS99] †[email protected] problem [Coo87]. Thus, as in other areas, the research focus has shifted from finding exact algorithms, towards finding good approximation schemes. Nevertheless there are important special instances of graphical models, e.g. trees, where exact algorithms are efficient. And, as we will see, even in the framework of variational methods, we want to use e...

Connections between Neural Networks and Boolean Functions∗

This report surveys some connections between Boolean functions and artificial neural networks. The focus is on cases in which the individual neurons are linear threshold neurons, sigmoid neurons, polynomial threshold neurons, or spiking neurons. We explore the relationships between types of artificial neural network and classes of Boolean function. In particular, we investigate the type of Bool...

Technical Note on Equivalence Between Recurrent Neural Network Time Series Models and Variational Bayesian Models

Abstract We observe that the standard log likelihood training objective for a Recurrent Neural Network (RNN) model of time series data is equivalent to a variational Bayesian training objective, given the proper choice of generative and inference models. This perspective may motivate extensions to both RNNs and variational Bayesian models. We propose one such extension, where multiple particles...