Raffaele Calabretta

Duplication of modules facilitates the evolution of functional specialization

The evolution of simulated robots with three different architectures is studied in this article. We compare a nonmodular feed-forward network, a hardwired modular, and a duplication-based modular motor control network. We conclude that both modular architectures outperform the non-modular architecture, both in terms of rate of adaptation as well as the level of adaptation achieved. The main difference between the hardwired and duplication-based modular architectures is that in the latter the modules reached a much higher degree of functional specialization of their motor control units with regard to high-level behavioral functions. The hardwired architectures reach the same level of performance, but have a more distributed assignment of functional tasks to the motor control units. We conclude that the mechanism through which functional specialization is achieved is similar to the mechanism proposed for the evolution of duplicated genes. It is found that the duplication of multifunctional modules first leads to a change in the regulation of the module, leading to a differentiation of the functional context in which the module is used. Then the module adapts to the new functional context. After this second step the system is locked into a functionally specialized state. We suggest that functional specialization may be an evolutionary absorption state.

Evolving Modular Architectures for Neural Networks

Neural networks that learn the What and Where task perform better if they possess a modular architecture for separately processing the identity and spatial location of objects. In previous simulations the modular architecture either was hardwired or it developed during an individual’s life based on a preference for short connections given a set of hardwired unit locations. We present two sets of simulations in which the network architecture is genetically inherited and it evolves in a population of neural networks in two different conditions: (1) both the architecture and the connection weights evolve; (2) the network architecture is inherited and it evolves but the connection weights are learned during life. The best results are obtained in condition (2). Condition (1) gives unsatisfactory results because (a) adapted sets of weights can suddenly become maladaptive if the architecture changes, (b) evolution fails to properly assign computational resources (hidden units) to the two tasks, (c) genetic linkage between sets of weights for different modules can result in a favourable mutation in one set of weights being accompanied by an unfavourable mutation in another set of weights.

What does it take to evolve behaviorally complex organisms

What genotypic features explain the evolvability of organisms that have to accomplish many different tasks? The genotype of behaviorally complex organisms may be more likely to encode modular neural architectures because neural modules dedicated to distinct tasks avoid neural interference, i.e. the arrival of conflicting messages for changing the value of connection weights during learning. However, if the connection weights for the various modules are genetically inherited, this raises the problem of genetic linkage: favorable mutations may fall on one portion of the genotype encoding one neural module and unfavorable mutations on another portion encoding another module. We show that this can prevent the genotype from reaching an adaptive optimum. This effect is different from other linkage effects described in the literature and we argue that it represents a new class of genetic constraints. Using simulations we show that sexual reproduction can alleviate the problem of genetic linkage by recombining separate modules all of which incorporate either favorable or unfavorable mutations. We speculate that this effect may contribute to the taxonomic prevalence of sexual reproduction among higher organisms. In addition to sexual recombination, the problem of genetic linkage for behaviorally complex organisms may be mitigated by entrusting evolution with the task of finding appropriate modular architectures and learning with the task of finding the appropriate connection weights for these architectures.

Two is better than one: a diploid genotype for neural networks

In nature the genotype of many organisms exhibits diploidy, i.e., it includes two copies of every gene. In this paper we describe the results of simulations comparing the behavior of haploid and diploid populations of ecological neural networks living in both fixed and changing environments. We show that diploid genotypes create more variability in fitness in the population than haploid genotypes and buffer better environmental change; as a consequence, if one wants to obtain good results for both average and peak fitness in a single population one should choose a diploid population with an appropriate mutation rate. Some results of our simulations parallel biological findings.

Genetic interference reduces the evolvability of modular and non-modular visual neural networks.

The aim of this paper is to propose an interdisciplinary evolutionary connectionism approach for the study of the evolution of modularity. It is argued that neural networks as a model of the nervous system and genetic algorithms as simulative models of biological evolution would allow us to formulate a clear and operative definition of module and to simulate the different evolutionary scenarios proposed for the origin of modularity. I will present a recent model in which the evolution of primate cortical visual streams is possible starting from non-modular neural networks. Simulation results not only confirm the existence of the phenomenon of neural interference in non-modular network architectures but also, for the first time, reveal the existence of another kind of interference at the genetic level, i.e. genetic interference, a new population genetic mechanism that is independent from the network architecture. Our simulations clearly show that genetic interference reduces the evolvability of visual neural networks and sexual reproduction can at least partially solve the problem of genetic interference. Finally, it is shown that entrusting the task of finding the neural network architecture to evolution and that of finding the network connection weights to learning is a way to completely avoid the problem of genetic interference. On the basis of this evidence, it is possible to formulate a new hypothesis on the origin of structural modularity, and thus to overcome the traditional dichotomy between innatist and empiricist theories of mind.

Diploid Robots Adapting to Fast Changing Environments

In most work applying genetic algorithms to populations of neural networks there is no real distinction between genotype and phenotype. In nature both the information contained in the genotype and the mapping of the genetic information into the phenotype are usually much more complex. Moreover, the genotypes of many organisms exhibit diploidy, i.e., they include two copies of each gene whose expression is governed by some dominance rules. We briefly review the literature on diploidy and we present our own model which in the present paper is applied to populations of organisms living in a fast changing environment Our results show that diploidy produce better performance than haploidy in this type of environments.

Adaptive Agents in Changing Environments, the Role of Modularity

We explored the role of modularity as a means to improve evolvability in populations of adaptive agents. We performed two sets of artificial life experiments. In the first, the adaptive agents were neural networks controlling the behavior of simulated garbage collecting robots, where modularity referred to the networks architectural organization and evolvability to the capacity of the population to adapt to environmental changes measured by the agents performance. In the second, the agents were programs that control the changes in network’s synaptic weights (learning algorithms), the modules were emerged clusters of symbols with a well defined function and evolvability was measured through the level of symbol diversity across programs. We found that the presence of modularity (either imposed by construction or as an emergent property in a favorable environment) is strongly correlated to the presence of very fit agents adapting effectively to environmental changes. In the case of learning algorithms we also observed that character diversity and modularity are also strongly correlated quantities.

Cellular Automata in an Artificial Life perspective

Artificial Life (AL; [1]) studies all kinds of biological phenomena as they occur in artificial organisms. Neural Networks (NNs; [2]) are commonly used in AL research as computational models of nervous systems that control organisms’ behavior. However, organisms do not only possess nervous systems and other phenotypic traits but also genetic information stored in the nucleus of their cells (genotype). The nervous system is part of the phenotype which is derived from this genotype through a process called development. The information specified in the genotype determines aspects of the nervous system which are expressed as innate behavioral tendencies and predisposition to learn. As a consequence, in AL models NNs tend to be accompanied by genotypes (i.e., genetic algorithms; [3]) and to become members of evolving populations of networks in which genotypes are inherited from parents to offspring.

Supporting argumentation in online political debate: Evidence from an experiment of collective deliberation:

In this article, we report the results of an e-democracy experiment in which a group of supporters of a large political party were asked to debate online about ways to reform the electoral law. We compare a traditional forum with an online collaborative argumentation platform to capture the various proposals and their associated pros and cons. The aim of this study is to assess the capability of this tool to support online collective deliberation in a real-world case, as compared to an online discussion supported by a forum. By comparing users’ experience across several metrics related to usability, activity levels, and quality of collaboration, our findings show that the forum produced more activity and ideas and its users perceived a better quality of the collaboration process, while the argumentation tool helped to reduce the amount of self-referential arguments and encourage viewing and rating of others’ posts.

Doparies: A Novel Party Deliberative and Aggregative Decision-Making Mechanism to Improve the Quality of Representative Democracy

The failings of parties are one of the central problems of contemporary democracies. What can be done to revive citizen participation? In this article, we present a novel party participatory decision-making mechanism named “doparies”. They are procedures that are nationally or locally implemented within and by parties, and permit any voter who declares to be an elector of that party (open doparies) or party members (internal doparies) to vote regarding crucial and controversial decisions during the period between one election and another. Whereas primaries are done before elections for choosing party candidates, doparies are done after elections for making party choices on issues. Doparies represent a bidirectional communication system between voters and representatives, and would retain the advantages of primaries (party–voters relationship) and referenda (debate before the vote), but would limit the excessive personalization of politics focusing on issues and not on people. There are both propositional doparies, allowing citizens to raise problems that are absent from their party political agenda, and consultative ones, allowing parties to hear the true voice of their voters, who, differently from what happens in polls, are informed by debates in party circles. We suggest that doparies are a new combination of deliberative and aggregative processes, and hypothesize that they can counteract parties’ crisis and abstention. Procedures similar to doparies are now part of the Italian Democratic Party statute and prominent national leaders have gathered signatures to organize local consultations. The use of primaries by Italian left-wing parties has had a contagious effect on right-wing ones as well as European ones. The same could happen with doparies.The failings of parties are one of the central problems of contemporary democracies. What can be done to revive citizen participation? In this article, we present a novel party participatory decision-making mechanism named “doparies”. They are procedures that are nationally or locally implemented within and by parties, and permit any voter who declares to be an elector of that party (open doparies) or party members (internal doparies) to vote regarding crucial and controversial decisions during the period between one election and another. Whereas primaries are done before elections for choosing party candidates, doparies are done after elections for making party choices on issues. Doparies represent a bidirectional communication system between voters and representatives, and would retain the advantages of primaries (party–voters relationship) and referenda (debate before the vote), but would limit the excessive personalization of politics focusing on issues and not on people. There are both propositional doparies, allowing citizens to raise problems that are absent from their party political agenda, and consultative ones, allowing parties to hear the true voice of their voters, who, differently from what happens in polls, are informed by debates in party circles. We suggest that doparies are a new combination of deliberative and aggregative processes, and hypothesize that they can counteract parties’ crisis and abstention. Procedures similar to doparies are now part of the Italian Democratic Party statute and prominent national leaders have gathered signatures to organize local consultations. The use of primaries by Italian left-wing parties has had a contagious effect on right-wing ones as well as European ones. The same could happen with doparies.