Simon Perkins

Supervisors: (2009)

Jeffrey Davies, Simon Perkins, Patrick Marais

There are many applications that require rendered 3D terrain, whether it be computer games or complex geographical research. The methods used to represent terrain, generate realistic terrain and...

A Spatial Awareness Framework for Enhancing Game Agent Behaviour (2008)

Perkins, Simon, Jacka, David, Marais, Patrick, Gain, James

We describe a framework for providing game agents with awareness of the intrinsic spatial qualities of the virtual worlds that they inhabit. We develop a novel data structure based on a modified...

Web-based design and evaluation of T-cell vaccine candidates (2008)

Thurmond, James, Yoon, Hyejin, Kuiken, Carla, Yusim, Karina, Perkins, Simon, Theiler, James, ...

Summary: We present a suite of on-line tools to design candidate vaccine proteins, and to assess antigen potential, using coverage of k-mers (as proxies for potential T-cell epitopes) as a metric....

processing (2007)

Steven P. Brumby, Neal R. Harvey, Simon Perkins, Reid B. Porter, John J. Szymanski, James Theiler, ...

genetic algorithm for combining new and existing image

Optimizing Digital Hardware Perceptrons for Multi-Spectral Image Classification (2007)

Reid Porter, Neal Harvey, Simon Perkins, James Theiler, Steven Brumby, Jeff Bloch, ...

Abstract. We propose a system for solving pixel-based multi-spectral image classification problems with high throughput pipelined hardware. We introduce a new shared weight network architecture that...

TASKS (2007)

Simon Perkins

Shaping is a way in which a human designer can provide assistance to a learning system to enable it to solve problems that would otherwise defeat it. Results are presented showing that shaping can...

1.1 From Robot Learning... (2007)

Simon Perkins, Gillian Hayes

Shaping is a way in which a human designer can provide assistance to a learning system to enable it to solve problems that would otherwise defeat it. The experiments reported here explore a variety...

H E UNIVER S I (2007)

Simon Perkins

ii In recent years, learning and evolutionary methods have been proposed as methods for automatically designing robot controllers without the need for detailed human design effort. Unfortunately, the...

Supervisors: (2007)

Jeffrey Davies, Simon Perkins, Patrick Marais

This project attempts to demonstrate that through creating an environment whereby virtual agents can easily gather information about their surroundings, more interesting behaviors can be created....

1. Project Description Terrain Evaluation for more effective AI (2007)

Jeff Davies, Paul Stephenson, Patrick Marais, Simon Perkins

The purpose of the project is to attempt to create agents, who have differing strengths and weaknesses, that can take advantage of terrain features to improve their effectiveness in accomplishing a...

Identification and reconstruction of bullets from multiple X-rays (2006)

Perkins, Simon, Marais, Patrick

We present a framework for the rapid detection and 3D localisation of bullets (or other compact shapes) from a sparse set of cross-sectional patient x-rays. The intention of this work is to assess a...

Proposed framework for anomalous change detection (2006)

James Theiler, Simon Perkins

For the anomalous change detection problem, you have a pair of images, taken of the same scene, but at different times and typically under different viewing conditions. You are looking for...

Grammar-guided feature extraction for time series classification. NIPS ‘05. Fast Time Series Classification Using Numerosity Reduction Random (2005)

Damian Eads, Karen Glocer, Simon Perkins, James Theiler

We present a flexible, general-purpose technique for generating time series classifiers. These classifiers are two-stage algorithms; each consists of a set of feature extraction programs, used for...

Identification and Reconstruction of Bullets from Multiple X-Rays (2004)

Perkins, Simon

The 3D shape and position of objects inside the human body are commonly detected using Computed Tomography (CT) scanning. CT is an expensive diagnostic option in economically disadvantaged areas and...

Grafting: Fast, incremental feature selection by gradient descent in function space (2003)

Simon Perkins, Kevin Lacker, James Theiler, Isabelle Guyon, André Elisseeff

We present a novel and flexible approach to the problem of feature selection, called grafting. Rather than considering feature selection as separate from learning, grafting treats the selection of...

Online feature selection using grafting (2003)

Simon Perkins, James Theiler

In the standard feature selection problem, we are given a fixed set of candidate features for use in a learning problem, and must select a subset that will be used to train a model that is “as good...

Grafting: Fast, incremental feature selection by gradient descent in function space (2003)

Simon Perkins, Kevin Lacker, James Theiler, Isabelle Guyon, André Elisseeff

We present a novel and flexible approach to the problem of feature selection, called grafting. Rather than considering feature selection as separate from learning, grafting treats the selection of...

Journal of Machine Learning Research 3 (2003) 1333-1356 Submitted 5/02; Published 3/03 Grafting: Fast, Incremental Feature Selection by (2003)

Gradient Descent In, Simon Perkins, Kevin Lacker, James Theiler, Isabelle Guyon, Andre Elisseeff

We present a novel and flexible approach to the problem of feature selection, called grafting. Rather than considering feature selection as separate from learning, grafting treats the selection of...

Two (2002)

Junshui Ma, James Theiler, Simon Perkins

realizations of a general feature extraction framework

Image Feature Extraction: GENIE vs Conventional Supervised Classification Techniques", accepted by IEEE Transactions on Geoscience and Remote Sensing (2002)

Neal R. Harvey, Steven P. Brumby, Simon Perkins, James Theiler, John J. Szymanski, J. Bloch, ...

Abstract — We have developed an automated feature detection/classification system, called Genie (GENetic Imagery Exploitation), which has been designed to generate image processing pipelines for a...

Comparison of GENIE and Conventional Supervised Classifiers for Multispectral Image Feature Extraction (2002)

Neal R. Harvey, James Theiler, Steven P. Brumby, Simon Perkins, John J. Szymanski, Jeffrey J. Bloch, ...

We have developed an automated feature detection /classification system, called GENetic Imagery Exploitation (GENIE), which has been designed to generate image processing pipelines for a variety of...

Support Vector Machines for Broad Area Feature Extraction (2001)

Simon Perkins, Neal R. Harvey, Steven P. Brumby, Kevin Lacker

Classification of broad area features in satellite imagery is one of the most important applications of remote sensing. It is often di#cult and time-consuming to develop classifiers by hand, so many...

A stateless client for progressive view-dependent transmission (2001)

Richard Southern, Simon Perkins, Barry Steyn, Alan Muller, Patrick Marais, Edwin Blake

We present a framework for real-time view-dependent refinement, and adapt it to the task of browsing large model repositories. We introduce a novel hierarchical representation of atomic operations...

Everything on the chip: a hardware-based self-contained spatially-structured genetic algorithm for signal processing (2000)

Simon Perkins, Reid Porter, Neal Harvey

Abstract. Evolutionary algorithms are useful optimization tools but are very time consuming to run. We present a self-contained FPGAbased implementation ofa spatially-structured evolutionary...

Everything on the chip: a hardware-based self-contained spatially-structured genetic algorithm for signal processing (2000)

Simon Perkins, Reid Porter, Neal Harvey

Abstract. Evolutionary algorithms are useful optimization tools but are very time consuming to run. We present a self-contained FPGAbased implementation of a spatially-structured evolutionary...

GENIE: A Hybrid Genetic Algorithm for Feature Classification (2000)

Simon Perkins, James Theiler, Steven P. Brumby, Neal R. Harvey, Reid Porter, John J. Szymanski, ...

We consider the problem of pixel-by-pixel classification of a multi-spectral image using supervised learning. Conventional supervised classification techniques such as maximiun likelihood...

Finding golf courses: The ultra high tech approach (2000)

Neal R. Harvey, Simon Perkins, Steven P. Brumby, James Theiler, Reid B. Porter, A. Cody Young, ...

Abstract. The search for a suitable golf course is a very important issue in the travel plans of any modern manager. Modern management is also infamous for its penchant for high-tech gadgetry. Here...

Internet Cookies: Security Implications (2000)

Simon Perkins

Cookies are an extension to the HTTP protocol. They provide state to the web browsing process. Cookies are actually an entry in the header of an HTTP request. A web server sends a cookie to a...

Evolving robot vision: Increasing performance through shaping (1999)

Simon Perkins

Abstract- Automated methods for designing robot controllers based on machine-learning techniques have shown great promise when applied to simple robot tasks, but in order to `scale up ' to more...

Evolving retrieval algorithms with a genetic programming scheme (1999)

James Theiler, Neal R. Harvey, Steven P. Brumby, John J. Szymanski, Steven Alferink, Simon Perkins, ...

The retrieval of scene properties (surface temperature, material type, vegetation health, etc.) from remotely sensed data is the ultimate goal of many earth observing satellites. The algorithms that...