Distributed Algorithms for Optimal Decision-Making
What is DiODe?
Consensus decisions, in which all members of a group contribute to making a single collective decision, are ubiquitous in biology, society, and engineering. This project aims towards a unified theory of collective consensus decision-making in biological and artificial systems.
The breakthrough insight is that in all physical decision-making systems decisions are reached by populations of components interacting with each other, and that it is not the nature of the components themselves but rather their patterns of interaction that are crucial.
Thus apparently very different decision-making systems, such as a social insect colony or a primate brain for example, may actually implement very similar mechanisms, one using neurons as interacting components and the other using ants or honeybees. Similarities in apparently different decision-making systems are not simply limited to brains and insect colonies, however.
This project is continuing recent work to develop the general theory of statistically-optimal decision-making strategies and value-sensitive collective decision strategies, seeking common mechanisms that instantiate them in living systems at diverse levels of biological complexity, from single cells, through primate brains, up to collectives of social individuals such as social insects or human groups.
The project will further develop this approach by theoretically and empirically studying the importance of both sampling strategies and individual confidence in collective consensus decision-making.
Finally, the project will also apply collective robotics with hundreds of micro-robots as a testbed for the theory developed, and as a future technology for solving complex collective decision tasks in a completely decentralised, yet statistically optimal, manner.
This interdisciplinary project aims to revolutionise the state-of-the-art in consensus decision-making, through identifying and applying novel optimality theory, and thereby advance and unify the study of decision-making in microbiology, entomology, psychology, neuroscience, robotics and political science.
As well as understanding the principles of collective decision-making and their application in artificial and natural systems, we aim to make our analytic and simulation tools freely available as intuitive open-source software, for engineers and biologists to use in modelling their own systems.
Marshall, J A R, Bogacz, R, Dornhaus, A, Planqué, R, Kovacs, T and Franks, N R (2009) ‘On optimal decision-making in brains and social insect colonies’, Journal of the Royal Society: Interface 6, pp.1065–1074.
Seeley, T D, Visscher, P K, Schlegel, T, Hogan, P M, Franks, N R and Marshall, J A R (2012) ‘Stop signals provide cross inhibition in collective decision-making by honeybee swarms’, Science 335, pp. 108–111.
Pais, D, Hogan, P M, Schlegel, T, Franks, N R, Leonard, N E and Marshall, J A R (2013) ‘A mechanism for value-sensitive decision-making’, PLoS one, e73216.
The DiODe project is funded by the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation programme (grant agreement number 647704).