CIGPU-2012 Australia, 10-15 June 2012

cigpu 2009 large logo GECCO 2009

GECCO 2009 Workshop, Tutorial and Competition on Computational Intelligence on Consumer Games and Graphics Hardware CIGPU-2009.

Room: St. Charles (Mezzanine floor) Delta Centre-Ville, 777 University Street, Montreal (airport bus, taxi $38.00 (CAD) one way (plus tip), Weather)
Paper submission deadline: March 25, 2009
Notification to authors: April 3, 2009
Submission of camera-ready version: April 17, 2009
Author Conference registration: April 27, 2009
Competition submission deadline: June 23, 2009

Programme Wednesday 8 July 2009

Time Title   Paper Photo
8:30 - 8:40 CIGPU 2009 Introduction Garnett Wilson pic
8:40 - 9:05 A Fast High Quality Pseudo Random Number Generator for nVidia CUDA Bill Langdon abs doi pic
9:05 - 9:30 Deployment of CPU and GPU-based Genetic Programming on Heterogenous Devices Garnett Wilson abs doi pic
9:30 - 9:55 Evolving Soft Robotic Locomotion in PhysX John Rieffel abs doi pic
9:55 - 10:20 Open discussion, including of experiences with latest GPU and commodity graphics technology Chaired by W. B. Langdon  
BREAK
10:40 - 11:05 Design and Implementation of Real-time Parallel GA Operators on the IBM Cell Processor Pascal Comte abs doi pic
11:05 - 11:30 Design and Implementation of Parallel Linear GP for the IBM Cell Processor Pascal Comte abs doi
11:30 - 11:55 Evaluating the Cell Broadband Engine as a Platform to Run Estimation of Distribution Algorithms Carlos Perez-Miguel abs doi pic
11:55 - 12:20 Solving Quadratic Assignment Problems by Genetic Algorithms with GPU Computation: A Case Study Shigeyoshi Tsutsui abs doi pic
LUNCH BREAK (including discussions)
14:00 - 14:25 Parallel Multi-Objective Evolutionary Algorithms on Graphics Processing Units Man Leung Wong abs doi pic
14:25 - 14:50 Parallel Latent Semantic Analyses using a Graphics Processing Unit Xiaohui Cui abs doi pic
14:50 - 15:50 Presentations by GPU competition entrants Chaired by Simon Harding  
BREAK 15:50 - 15:55 CIGPU 2010 planning W. B. Langdon  
16:10 - 18:00 GPU Tutorial, Cartier B (Floor C) doi Wolfgang Banzhaf and Simon Harding  

Publication

All accepted papers will be presented at CIGPU 2009 and published by ACM Press in the workshop proceedings to be included on the GECCO Proceedings and Companion Material CD and placed in the ACM Digital Library.

Workshop Format and Schedule

CIGPU 2009 will be held as a full day event (see above) including:

Workshop Chairs

Program Committee

Malcolm Heywood, Wai-Man Pang, Raghavendra D. Prabhu, Tien-Tsin Wong, Stephane Gobron, Ignacio Hidalgo, Francisco Fernandez de Vega, Juan Lanchares, Denis Robilliard

Background

Call for Papers (PDF)

The Second International Workshop and Tutorial on Computational Intelligence on Consumer Games and Graphics Hardware (CIGPU 2009) will be held in Montreal, Canada, as part of the Genetic and Evolutionary Computation Conference (GECCO-2009), July 8-12, 2009.
Everyone who has implemented an evolutionary algorithm or other computational intelligence approach using graphics processing units (GPUs), video game systems, mobile devices, cellular phones, etc. will want to submit to this workshop. Due to its speed, price, and availability, there is increasing interest in using mass consumer market commodity hardware for engineering and scientific applications. Mostly this has concentrated upon graphics hardware, particularly GPUs, due to their ability to offer teraflop performance on a desktop using a restricted form of parallel computing (known as General Purpose computing on Graphics Processing Units, or GPGPU). There is also increasing interest in using the computing power of game consoles such as Microsoft's Xbox, Sony's Playstation and the Cell processor, and portable entertainment and/or cellular phone mobile devices for research and applications.

Submissions were invited in (but not limited to) the following areas:

Aims and Objectives

GPUs provide a very restricted type of parallel processing. PC cards like 9800 GTX or RV670 aught to give terraflop at your desk but how to actually get hold of it for non-trivial applications? One of the principle benefits of the CIGPU special session will be to disseminate knowledge of tools and techniques that are required to get the best from these widely available but difficult to fully exploit consumer hardware.

Introduction

Due to its speed, price and availability, there is increasing interest in using mass consumer market commodity hardware for engineering and scientific applications. Mostly this has concentrated upon graphics hardware, particularly GPUs. However there is also increasing interest in using games consoles such as Microsoft's Xbox, Sony's Playstation and the Cell processor, for research and applications. Indeed, nVidia have recently announced Tesla as a "scientificated" version of their newest GPUs. In the future, personal computer physics engines, which are intended to provide realistic re al time simulation of multi-body physics for sophisticated games, may also be adapted to serve science (rather than simulate it). At WCCI 2008, a special session was held focusing on the use of these platforms for computational intelligence. CIGPU-2008 proved to be a successful event. We were not only able to select sufficient, high quality papers but also to have two panel discussions. A full review of the session can be found in the SIGEVO Spring 2008 newsletter. Given the success of CIGPU-2008, we are confident that a follow up workshop will also be well received. A list of publications related to evolutionary computation on GPU type platforms can be found at http://www.gpgpgpu.com.

Interest to the Community

The use of commodity graphics hardware for scientific computing (often referred to as General Purpose computing on Graphics Processing Units, GPGPU) is rapidly becoming established. However, their use in evolutionary computing is only emerging. Why is this? GPUs provide a very restricted type of parallel processing. PC cards like 8800 GTX or RV670 ought to give teraflop performance at your desk, but the question is how to actually harness it for non-trivial applications. One of the principle benefits of the CIGPU workshop will be to disseminate knowledge of tools and techniques that are required to get the best from these widely available, but difficult to fully exploit, consumer hardware platforms. The workshop will allow researchers to demonstrate the types of applications that can be implem ented on this form of hardware. We expect the computational power of these devices to enable significant advances in evolutionary computing to be made.

Other GPU sites

A few links to other pages interested in geneneral purpose computing on graphics hardware

Acknowledgments

CIGPU 2009 logo generously designed and contributed by Taras Kowaliw.

cigpu 2009 Garnett's logo cigpu 2009 old logo

W.B.Langdon 15 Oct 2008 (last update 2 July 2014)