Matthew Studley

Learning Classifier System Ensembles With Rule-Sharing

This paper presents an investigation into exploiting the population-based nature of learning classifier systems (LCSs) for their use within highly parallel systems. In particular, the use of simple payoff and accuracy-based LCSs within the ensemble machine approach is examined. Results indicate that inclusion of a rule migration mechanism inspired by parallel genetic algorithms is an effective way to improve learning speed in comparison to equivalent single systems. Presentation of a mechanism which exploits the underlying niche-based generalization mechanism of accuracy-based systems is then shown to further improve their performance, particularly, as task complexity increases. This is not found to be the case for payoff-based systems. Finally, considerably better than linear speedup is demonstrated with the accuracy-based systems on a version of the well-known Boolean logic benchmark task used throughout.

Is social norms marketing effective? A case study in domestic electricity consumption

Purpose – The purpose of this paper is to present details of a large-scale experiment that evaluated the impact of communicating two types of feedback to householders regarding their domestic electricity consumption: feedback on their own consumption and feedback of both their own consumption and that of others in their locality. Design/methodology/approach – Digital technologies were used to automatically measure and communicate the electricity consumption of 316 UK residents for a period of 16 weeks. Participants were randomly assigned to one of three experimental conditions: one involving no feedback; one involving feedback about a households own usage, and one involving a households own usage plus social norms feedback (the average consumption of others in the locality). At the end of the study, a selection of participants took part in interviews or focus groups. Findings – Both types of feedback (individual and individual-plus-social-norms) led to reductions in consumption of about 3 per cent. Thos...

X-TCS: accuracy-based learning classifier system robotics

Most research in the held of learning classifier systems today concentrates on the accuracy-based XCS. This paper presents initial results from an extension of XCS that operates in continuous environments on a physical robot. This is compared with a similar extension based upon the simpler ZCS. The new system is shown to be capable of near optimal performance in a simple robotic task. To the best of our knowledge, this is the first application of an accuracy-based LCS to controlling a physical agent in the real world without a priori discretization.

On the use of rule-sharing in learning classifier system ensembles

This paper presents an investigation into exploiting the population-based nature of learning classifier systems for their use within highly-parallel systems. In particular, the use of simple accuracy-based learning classifier systems within the ensemble machine approach is examined. Results indicate that inclusion of a rule migration mechanism inspired by parallel genetic algorithms is an effective way to improve learning speed

The distributed co-evolution of an on-board simulator and controller for swarm robot behaviours

Abstract We investigate the reality gap, specifically the environmental correspondence of an on-board simulator. We describe a novel distributed co-evolutionary approach to improve the transference of controllers that co-evolve with an on-board simulator. A novelty of our approach is the the potential to improve transference between simulation and reality without an explicit measurement between the two domains. We hypothesise that a variation of on-board simulator environment models across many robots can be competitively exploited by comparison of the real controller fitness of many robots. We hypothesise that the real controller fitness values across many robots can be taken as indicative of the varied fitness in environmental correspondence of on-board simulators, and used to inform the distributed evolution an on-board simulator environment model without explicit measurement of the real environment. Our results demonstrate that our approach creates an adaptive relationship between the on-board simulator environment model, the real world behaviour of the robots, and the state of the real environment. The results indicate that our approach is sensitive to whether the real behavioural performance of the robot is informative on the state real environment.

Using the XCS Classifier System for Multi-objective Reinforcement Learning Problems

We investigate the performance of a learning classifier system in some simple multi-objective, multi-step maze problems, using both random and biased action-selection policies for exploration. Results show that the choice of action-selection policy can significantly affect the performance of the system in such environments. Further, this effect is directly related to population size, and we relate this finding to recent theoretical studies of learning classifier systems in single-step problems.

Consideration of Multiple Objectives in Neural Learning Classifier Systems

For effective use in a number of problem domains Learning Classifier Systems must be able to manage multiple objectives. This paper explicitly considers the case of developing the controller for a simulated mobile autonomous robot which must achieve a given task whilst maintaining sufficient battery power. A form of Learning Classifier System in which each rule is represented by an artificial neural network is used. Results are presented which show it is possible to solve both objectives when the energy level is presented as an input along with sensor data. A more realistic, and hence more complex, version of the basic scenario is then investigated.

The distributed co-evolution of an embodied simulator and controller for swarm robot behaviours

Embodied fitness assessment of robotic controllers is slow but grounded, while assessment in a simulated environment is fast but can run foul of the ‘reality gap’. We present a distributed co-evolutionary method to adapt the environmental model of an on-board simulator within the context of swarm robotics.

Mobile GPGPU Acceleration of Embodied Robot Simulation

It is desirable for a robot to be able to run on-board simulations of itself in a model of the world to evaluate action consequences and test new controller solutions, but simulation is computationally expensive. Modern mobile System-on-Chip devices have high performance at low power consumption levels and now incorporate powerful graphics processing units, making them good potential candidates to host on-board simulations. We use the parallel language OpenCL on two such devices to accelerate the widely-used Stage robot simulator and demonstrate both higher simulation speed and lower energy use on a multi-robot benchmark. To the best of our knowledge, this is the first time that GPGPU on mobile devices have been used to accelerate robot simulation, and moves towards providing an autonomous robot with an embodied what-if capability.

Evolving behaviour trees for swarm robotics

Controllers for swarms of robots are hard to design as swarm behaviour emerges from their interaction, and so controllers are often evolved. However, these evolved controllers are often difficult to understand, limiting our ability to predict swarm behaviour. We suggest behaviour trees are a good control architecture for swarm robotics, as they are comprehensible and promote modular reuse. We design a foraging task for kilobots and evolve a behaviour tree capable of performing that task, both in simulation and reality, and show the controller is compact and understandable.