Channel Compression: Rethinking Information Redundancy Among Channels in CNN Architecture

Graph compression—save information by exploiting redundancy

In this paper we raise the question of how to compress sparse graphs. By introducing the idea of redundancy, we find a way to measure the overlap of neighbors between nodes in networks. We exploit symmetry and information by making use of the overlap in neighbors and analyzing how information is reduced by shrinking the network and, using the specific data structure we created, we generalize th...

Facets Are Fundamental: Rethinking Information Architecture Frameworks

FOR TECHNICAL COMMUNICATION ONLINE This article presents three problems with existing information architecture frameworks.First, they are too focused on organizing information based on topic. Second, they treatfacets as a supplemental form of classification. Third, they conflate the organization andrepresentation of information. Analysis of these three problems suggests thatinfo...

Image Compression by Cnn

The Opacity of Glass Rethinking Transparency in Contemporary Architecture

In this paper, an exchange of letters between the philosopher Jacques Derrida and architect Peter Eisenman is used as a catalyst to discuss the material qualities of glass and its relationship with the concept of transparency in architecture. In his criticism of an architecture devoid of human qualities, Derrida uses glass, defined through Walter Benjamin’s writings, as a hard and cold material...

Information Theory: Data Compression, Channel Coding, Entropy, and Mutual Information

This report gives an insight about two major aspects of information theory, including data compression and channel coding. It discusses the simulations results obtained using MATLAB from the information theory view point. It also defines these aspects by using the definitions of entropy and mutual information which reflects the optimum data compression could be obtained and the ultimate transmi...