Communications of the ACM
Latest Research
From Communications of the ACM
Technical Perspective: What Led Computer Vision to Deep Learning?
We are in the middle of the third wave of interest in artificial neural networks as the leading paradigm for machine learning. "ImageNet Classification with Deep...
From Communications of the ACM
ImageNet Classification with Deep Convolutional Neural Networks
In the 1980s backpropagation did not live up to the very high expectations of its advocates. Twenty years later, we know what went wrong: for deep neural networks...
From Communications of the ACM
Technical Perspective: Functional Compilers
"Exploiting Vector Instructions with Generalized Stream Fusion" points out that stream fusion by itself is not well suited for generating bulk instructions such...
From Communications of the ACM
Exploiting Vector Instructions with Generalized Stream Fusion
Programmers should not have to sacrifice code clarity or good software engineering practices to obtain performance. This work shows how to attain this goal for...
From Communications of the ACM
Technical Perspective: Data Distribution For Fast Joins
What is the most drastic way to reduce the cost of communication for parallel data processing algorithms? This is the question studied in "Reasoning on Data Partitioning...
From Communications of the ACM
Reasoning on Data Partitioning For Single-Round Multi-Join Evaluation in Massively Parallel Systems
We introduce a framework for reasoning about data partitioning to detect when we can avoid the data reshuffling step.
From Communications of the ACM
Authentication Using Pulse-Response Biometrics
We propose a new biometric based on the human body's response to an electric square pulse signal, called pulse-response.
From Communications of the ACM
Technical Perspective: Mapping the Universe
"HACC: Extreme Scaling and Performance Across Diverse Architectures" describes the Hardware/Hybrid Accelerated Cosmology Code (HACC) framework, which uses a novel...
From Communications of the ACM
HACC: Extreme Scaling and Performance Across Diverse Architectures
In this Research Highlight, we demonstrate the success of HACC on two very different machines, the CPU/GPU system Titan and the BG/Q systems Sequoia and Mira, attaining...
From Communications of the ACM
Technical Perspective: Jupiter Rising
As "Jupiter Rising" makes clear, many of the Internet mechanisms for maintaining large-scale networks are suboptimal when the datacenter is largely homogeneous,...
From Communications of the ACM
Jupiter Rising: A Decade of Clos Topologies and Centralized Control in Google's Datacenter Network
We present our approach for overcoming the cost, operational complexity, and limited scale endemic to datacenter networks a decade ago.
From Communications of the ACM
Technical Perspective: Mesa Takes Data Warehousing to New Heights
Producing reports at the scale of Google Ads, where billions of clicks happen per day, is the challenge addressed by the Mesa system described in "Mesa: A Geo-Replicated...
From Communications of the ACM
Mesa: A Geo-Replicated Online Data Warehouse For Google's Advertising System
Mesa is a highly scalable analytic data warehousing system that stores critical measurement data related to Google's Internet advertising business. This paper presents...
From Communications of the ACM
Technical Perspective: The State (and Security) of the Bitcoin Economy
"A Fistful of Bitcoins" examines, in the context of Bitcoin, what we could learn by studying the patterns encoded in a complete record of every single financial...
From Communications of the ACM
A Fistful of Bitcoins: Characterizing Payments Among Men with No Names
Bitcoin has the unintuitive property that while the ownership of money is implicitly anonymous, its flow is globally visible. In this paper we explore this unique...
From Communications of the ACM
Technical Perspective: Fairness and the Coin Flip
"Secure Multiparty Computations on Bitcoin" introduces an exciting new idea for how to provide fairness: leverage Bitcoin’s existing infrastructure for distributed...
From Communications of the ACM
Secure Multiparty Computations on Bitcoin
In this work, we propose to use Bitcoin to design fully decentralized protocols that are secure even if no trusted third party is available.
From Communications of the ACM
Technical Perspective: Taming the Name Game
In "Learning to Name Objects," the authors offer a method to determine a basic-level category name for an object in an image.
From Communications of the ACM
Learning to Name Objects
This paper looks at the problem of predicting category labels that mimic how human observers would name objects.
From Communications of the ACM