The present paper analyzes the mutual relationships between generative and developmental systems (GDS) and synaptic plasticity when evolving plastic artificial neural networks (ANNs) in reward-based scenarios. We first introduce the concept of synaptic Transitive Learning Abilities (sTLA), which reflects how well an evolved plastic ANN can cope with learning scenarios not encountered during the evolution process. We subsequently report results of a set of experiments designed to check that (1) synaptic plasticity can help a GDS to fine-tune synaptic weights and (2) that with the investigated generative encoding (EvoNeuro), only a few learning scenarios are necessary to evolve a general learning system, which can adapt itself to reward-based scenarios not tested during the fitness evaluation.

A major goal of bio-inspired artificial intelligence is to design artificial neural networks with abilities that resemble those of animal nervous systems. It is commonly believed that two keys for evolving nature-like artificial neural networks are (1) the developmental process that links genes to nervous systems, which enables the evolution of large, regular neural networks, and (2) synaptic plasticity, which allows neural networks to change during their lifetime. So far, these two topics have been mainly studied separately. The present paper shows that they are actually deeply connected. Using a simple operant conditioning task and a classic evolutionary algorithm, we compare three ways to encode plastic neural networks: a direct encoding, a developmental encoding inspired by computational neuroscience models, and a developmental encoding inspired by morphogen gradients (similar to HyperNEAT). Our results suggest that using a developmental encoding could improve the learning abilities of evolved, plastic neural networks. Complementary experiments reveal that this result is likely the consequence of the bias of developmental encodings towards regular structures: (1) in our experimental setup, encodings that tend to produce more regular networks yield networks with better general learning abilities; (2) whatever the encoding is, networks that are the more regular are statistically those that have the best learning abilities.

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Abstract-Many controllers for complex agents have been successfully generated by automatically desiging artificial neural networks with evolutionary algorithms. However, typical evolved neural networks are not able to adapt themselves online, making them unable to perform tasks that require online adaptation. Nature solved this problem on animals with plastic nervous systems. Inpired by neuroscience models of plastic neuralnetwork, the present contribution proposes to use a combination of Hebbian learning, neuro-modulation and a a generative mapbased encoding. We applied the proposed approach on a problem from operant conditioning (a Skinner box), in which numerous different association rules can be learned. Results show that the map-based encoding scaled up better than a classic direct encoding on this task. Evolving neural networks using a mapbased generative encoding also lead to networks that works with most rule sets even when the evolution is done on a small subset of all the possible cases. Such a generative encoding therefore appears as a key to improve the generalization abilities of evolved adaptive neural networks.