RULE 2008 Neural Networks in Maude ⋆ Gustavo Santos-Garc´ ıa 1 Universidad de Salamanca Miguel Palomino 2 Departamento de Sistemas Inform´aticos y Computaci´on, UCM Alberto Verdejo 3 Departamento de Sistemas Inform´aticos y Computaci´on, UCM Abstract In this work we study the representation of the computational model of artificial neural networks in rewriting logic, along the lines of several models of parallelism and concurrency that have already been mapped into it. We show how crucial is the right choice for the representation operations and the availability of strategies to guide the application of our rules. Finally, we also apply our specification to data used in the diagnosis of glaucoma. Keywords: Neural networks, rewriting logic, Maude, strategies, executability. 1 Introduction Rewriting logic [11] is a logic of concurrent change that can naturally deal with states and with highly nondeterministic concurrent computations. It has good properties as a flexible and general semantic framework for giving semantics to a wide range of languages and models of concurrency. Indeed, rewriting logic was proposed as a unifying framework in which many models of concurrency could be represented, such as labeled transition systems, phrase structure grammars, Petri nets, concurrent object-oriented programming, or CCS, to name a few. For many of these models, concrete maps have actually been defined into rewriting logic; see e.g. [15,14,12,7,16] and the references in [8]. ⋆ Research supported by Spanish project DESAFIOS TIN2006–15660–C02–01 and by Comunidad de Madrid program PROMESAS S–0505/TIC/0407. 1 Email:santos@usal.es 2 Email:miguelpt@sip.ucm.es 3 Email:alberto@sip.ucm.es This paper is electronically published in Electronic Notes in Theoretical Computer Science URL: www.elsevier.nl/locate/entcs