Communications of the ACM
Research Archive
The Research archive provides access to all Research articles published in past issues of Communications of the ACM.
October 2011
From Communications of the ACM
Technical Perspective: A Better Way to Learn Features
A typical machine learning program uses weighted combinations of features to discriminate between classes or to predict real-valued outcomes. The art of machine learning is…
From Communications of the ACM
Technical Perspective: Power Efficiency as the #1 Design Constraint
Moore's Law, and associated observations by Bob Dennard, describe key technical foundations of the semiconductor industry…
From Communications of the ACM
Technical Perspective: Visual Reconstruction
Nearly 460,000 Flickr pictures were used to create detailed three-dimensional geometry and colors of famous landmarks and monuments in Rome, Venice, and…
From Communications of the ACM
Unsupervised Learning of Hierarchical Representations with Convolutional Deep Belief Networks
There has been much interest in unsupervised learning of hierarchical generative models such as deep belief networks (DBNs); however, scaling such models to full-sized, high-dimensional images remains a difficult problem.
From Communications of the ACM
Understanding Sources of Ineffciency in General-Purpose Chips
To better understand what improvement in processor efficiency is possible, we quantify the performance and energy overheads of a 720p HD H.264 encoder running on a CMP system. We explore methods to eliminate these overheads by…
From Communications of the ACM
Building Rome in a Day
We present a system that can reconstruct 3D geometry from large, unorganized collections of photographs. Our experimental results demonstrate that it is possible to reconstruct city-scale image collections with more than a hundred…