Convolutional neural networks (CNN) are increasingly used in many areas of computer vision. They are particularly attractive because of their ability to “absorb” great quantities of labeled data through millions of parameters. However, as model sizes increase, so do the storage and memory requirements of the classifiers. We present a novel network architecture, Frequency-Sensitive Hashed Nets (...

Convoulutional Neural Networks (CNNs) exhibit extraordinary performance on a variety of machine learning tasks. However, their mathematical properties and behavior are quite poorly understood. There is some work, in the form of a framework, for analyzing the operations that they perform. The goal of this project is to present key results from this theory, and provide intuition for why CNNs work.

Convolutional neural networks have recently been shown to outperform fully connected deep neural networks on several speech recognition tasks. Their superior performance is due to their convolutional structure that processes several, slightly shifted versions of the input window using the same weights, and then pools the resulting neural activations. This pooling operation makes the network les...

We present a new deterministic relational model derived from convolutional neural networks. Search-Convolutional Neural Networks (SCNNs) extend the notion of convolution to graph search to construct a rich latent representation that extracts local behavior from general graph-structured data. Unlike other neural network models that take graph-structured data as input, SCNNs have a parameterizati...

6 The convolution layer 13 6.1 What is a convolution? . . . . . . . . . . . . . . . . . . . . . . . . 13 6.2 Why to convolve? . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 6.3 Convolution as matrix product . . . . . . . . . . . . . . . . . . . 18 6.4 The Kronecker product . . . . . . . . . . . . . . . . . . . . . . . 20 6.5 Backward propagation: update the parameters . . . . . . . . ...

Neural networks have recently had a lot of success for many tasks. However, neural network architectures that perform well are still typically designed manually by experts in a cumbersome trial-and-error process. We propose a new method to automatically search for well-performing CNN architectures based on a simple hill climbing procedure whose operators apply network morphisms, followed by sho...

We present diffusion-convolutional neural networks (DCNNs), a new model for graph-structured data. Through the introduction of a diffusion-convolution operation, we show how diffusion-based representations can be learned from graphstructured data and used as an effective basis for node classification. DCNNs have several attractive qualities, including a latent representation for graphical data ...

We present a simple approach based on pixel-wise nearest neighbors to understand and interpret the internal operations of state-of-the-art neural networks for pixel-level tasks. Specifically, we aim to understand the synthesis and prediction mechanisms of state-of-the-art convolutional neural networks for pixel-level tasks. To this end, we primarily analyze the synthesis process of generative m...