Research paves the way for unselfish computerised agents
A scientist from the University of Southampton will today (Thursday 19 August) unveil a new computer simulation that could explain how selfish agents learn to cooperate in nature.
Adam Davies, a PhD student at the University's School of Electronics and Computer Science (ECS) will present a paper at the annual Artificial Life XII - 12th International Conference on the Synthesis and Simulation of Living Systems in Denmark on Thursday 19 August. Adam will reveal a new adaptive system that makes it possible to encourage agents to change their purely selfish behaviours and create a scenario where although the agents are still selfish, their actions favour the common good. In his paper entitled "If you can't be with the one you love, love the one you're with: How individual habituation of agent interactions improves global utility", Adam describes how it is possible to draw on evolutionary biology to get self-interested agents to modify their behaviour. The approach is based on using Hebbian learning, a simple learning process that occurs in the brain, to develop agents that behave as creatures of habit and can learn to make decisions that in the long run maximise global utility. "Our research looks at the effect of a strategy for increasing total utility in systems of selfish agents. The strategy is simple – we just let agents learn to prefer doing things they have done before," says Adam, who is doing a PhD on evolutionary simulation at the Institute for Complex Systems Simulation (ICSS).
"With this strategy in place, selfish agents make decisions based on their learnt preferences rather than their true utility. Amazingly, doing this in the long term causes agents to coordinate their behaviours and maximise global utility. Basically it enables cooperation, making everyone better off. This work thereby helps us to understand self-organisation in nature, and sheds light on how to devise simple strategies that sidestep selfishness.”
Adam Davies is a member of the Institute for Complex Systems Simulation. If you are interested in doing PhD research in this group you can find out more on our Postgraduate Admissions Pages.
For further information about this news story contact Joyce Lewis; tel.+44(0)23 8059 5453.