Constraining connectivity to encourage modularity in HyperNEAT

A challenging goal of generative and developmental systems (GDS) is to effectively evolve neural networks as complex and capable as those found in nature. Two key properties of neural structures in nature are regularity and modularity. While HyperNEAT has proven capable of generating neural network connectivity patterns with regularities, its ability to evolve modularity remains in question. This paper investigates how altering the traditional approach to determining whether connections are expressed in HyperNEAT influences modularity. In particular, an extension is introduced called a Link Expression Output (HyperNEAT-LEO) that allows HyperNEAT to evolve the pattern of weights independently from the pattern of connection expression. Because HyperNEAT evolves such patterns as functions of geometry, important general topographic principles for organizing connectivity can be seeded into the initial population. For example, a key topographic concept in nature that encourages modularity is locality, that is, components of a module are located near each other. As experiments in this paper show, by seeding HyperNEAT with a bias towards local connectivity implemented through the LEO, modular structures arise naturally. Thus this paper provides an important clue to how an indirect encoding of network structure can be encouraged to evolve modularity.

[1]  W. J. Freeman,et al.  Alan Turing: The Chemical Basis of Morphogenesis , 1986 .

[2]  A. M. Turing,et al.  The chemical basis of morphogenesis , 1952, Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences.

[3]  L. Altenberg,et al.  PERSPECTIVE: COMPLEX ADAPTATIONS AND THE EVOLUTION OF EVOLVABILITY , 1996, Evolution; international journal of organic evolution.

[4]  J. Hopfield,et al.  From molecular to modular cell biology , 1999, Nature.

[5]  Risto Miikkulainen,et al.  Evolving Neural Networks through Augmenting Topologies , 2002, Evolutionary Computation.

[6]  Risto Miikkulainen,et al.  A Taxonomy for Artificial Embryogeny , 2003, Artificial Life.

[7]  Annie S. Wu,et al.  Effects of Module Encapsulation in Repetitively Modular Genotypes on the Search Space , 2004, GECCO.

[8]  Risto Miikkulainen,et al.  Competitive Coevolution through Evolutionary Complexification , 2011, J. Artif. Intell. Res..

[9]  Shimon Whiteson.  Improving reinforcement learning function approximators via neuroevolution , 2005, AAMAS '05.

[10]  U. Alon,et al.  Spontaneous evolution of modularity and network motifs. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[11]  Shimon Whiteson,et al.  Comparing evolutionary and temporal difference methods in a reinforcement learning domain , 2006, GECCO.

[12]  Kenneth O. Stanley,et al.  Generating large-scale neural networks through discovering geometric regularities , 2007, GECCO '07.

[13]  Kenneth O. Stanley,et al.  A novel generative encoding for exploiting neural network sensor and output geometry , 2007, GECCO '07.

[14]  Shimon Whiteson,et al.  Stochastic Optimization for Collision Selection in High Energy Physics , 2006, AAAI.

[15]  Hod Lipson,et al.  Principles of modularity, regularity, and hierarchy for scalable systems , 2007 .

[16]  Kenneth O. Stanley,et al.  Compositional Pattern Producing Networks : A Novel Abstraction of Development , 2007 .

[17]  Kenneth O. Stanley,et al.  A Case Study on the Critical Role of Geometric Regularity in Machine Learning , 2008, AAAI.

[18]  Kenneth O. Stanley,et al.  Generative encoding for multiagent learning , 2008, GECCO '08.

[19]  Charles Ofria,et al.  Evolving coordinated quadruped gaits with the HyperNEAT generative encoding , 2009, 2009 IEEE Congress on Evolutionary Computation.

[20]  Kenneth O. Stanley,et al.  A Hypercube-Based Encoding for Evolving Large-Scale Neural Networks , 2009, Artificial Life.

[21]  Kenneth O. Stanley.  A Hypercube-Based Indirect Encoding for Evolving Large-Scale Neural Networks , 2009 .

[22]  Annie S. Wu,et al.  On the performance effects of unbiased module encapsulation , 2009, GECCO '09.

[23]  Kenneth O. Stanley,et al.  Autonomous Evolution of Topographic Regularities in Artificial Neural Networks , 2010, Neural Computation.

[24]  Joel Lehman,et al.  Evolving policy geometry for scalable multiagent learning , 2010, AAMAS.

[25]  Kenneth O. Stanley,et al.  Evolving Static Representations for Task Transfer , 2010, J. Mach. Learn. Res..

[26]  Charles Ofria,et al.  Investigating whether hyperNEAT produces modular neural networks , 2010, GECCO '10.

[27]  Kenneth O. Stanley,et al.  On the Performance of Indirect Encoding Across the Continuum of Regularity , 2011, IEEE Transactions on Evolutionary Computation.