Interview with Yves Duthen, designing artificial systems from the living

We are publishing the fifth video in a series of interviews presenting the research work of our different departments. Yves DUTHEN, teacher-researcher at department CISO – Team REVA, explains his research work on artificial life. It is a field that combines computer science, biology and neuroscience in order to design artificial systems inspired by living organisms, and to provide tools to explore and understand living organisms.

Artificial life, when computing is inspired by life

Artificial life is interested in the properties of the living world, with the aim of designing artificial systems that reproduce its characteristics. Among these biological properties that interest computer science researchers, we can cite simulated evolution taken up by algorithms that hybridize programs; ontogeny, which consists of “growing a system”; adaptation and learning, which are the inspiration for artificial intelligence; and replication, as in cell division, for example. Research on artificial life has been developing over the last thirty years, starting with initial work in behavioral simulation and form synthesis. The ambition is to draw inspiration from living organisms to design new computer systems (in particular resilient ones) and also to imagine virtual models (in silico) to simulate living organisms in a multidisciplinary approach. The increase in knowledge in biology associated with the advances made in computing power allows life sciences to be equipped to, as Yves Duthen says, “grasp the intimacy of life.” In the field of computer science, research on artificial life concerns three concrete fields of application: software (algorithms inspired by life), hardware (robotics) and wetware (synthetic biology). 

What are the challenges in this area of research? 

“Some researchers believe that today, in the field of synthetic biology, we are in the same position as in 1970 in the field of computer science, when tools were available to manipulate only a few bytes.” Yves DUTHEN

Today, gene editing and protein synthesis tools make it possible to manipulate and synthesize the elementary building blocks of living organisms, allowing the possibility of programming bacteria or cells to perform functions (DNA compiler). The possibility to develop systems, to replicate them or to hybridize them with living cells would allow numerous applications in the medical and environmental fields but with immense security problems.

On the other hand, concerning artificial intelligence, there is a huge gap between machine learning and human cognition (understanding the world). The question that preoccupies researchers is to understand what are the representations of the world and the learning mechanisms of living beings, cells, plants and evolved organisms, and if these are replicable in artificial systems. Hybridizing artificial, living and cognitive systems obviously raises major ethical questions, which make this field a crossroads where different scientific and human disciplines meet.