A modular architecture for transparent computation in recurrent neural networks

Abstract : Computation is classically studied in terms of automata, formal languages and algorithms; yet, the relation between neural dynamics and symbolic representations and operations is still unclear in traditional eliminative connectionism. Therefore, we suggest a unique perspective on this central issue, to which we would like to refer as to transparent connectionism, by proposing accounts of how symbolic computation can be implemented in neural substrates. In this study we first introduce a new model of dynamics on a symbolic space, the versatile shift, showing that it supports the real-time simulation of a range of automata. We then show that the Gödelization of versatile shifts defines nonlinear dynamical automata, dynamical systems evolving on a vectorial space. Finally, we present a mapping between nonlinear dynamical automata and recurrent artificial neural networks. The mapping defines an architecture characterized by its granular modularity, where data, symbolic operations and their control are not only distinguishable in activation space, but also spatially localizable in the network itself, while maintaining a distributed encoding of symbolic representations. The resulting networks simulate automata in real-time and are programmed directly, in absence of network training. To discuss the unique characteristics of the architecture and their consequences, we present two examples: i) the design of a Central Pattern Generator from a finite-state locomotive controller, and ii) the creation of a network simulating a system of interactive automata that supports the parsing of garden-path sentences as investigated in psycholinguistics experiments.
Liste complète des métadonnées

https://hal.inria.fr/hal-01386281
Contributeur : Mathieu Desroches <>
Soumis le : dimanche 23 octobre 2016 - 16:12:58
Dernière modification le : jeudi 11 janvier 2018 - 16:17:00

Identifiants

Collections

Citation

Giovanni Carmantini, Peter Beim Graben, Mathieu Desroches, Serafim Rodrigues. A modular architecture for transparent computation in recurrent neural networks. Neural Networks, Elsevier, 2017, 85 (1), pp.85-105. 〈http://www.sciencedirect.com/science/article/pii/S0893608016301198〉. 〈10.1016/j.neunet.2016.09.001〉. 〈hal-01386281〉

Partager

Métriques

Consultations de la notice

178