Complex genetic evolution of artificial self‐replicators in cellular automata

Abstract

It is widely believed that evolutionary dynamics of artificial self-replicators realized in cellular automata (CA) Key Words: cellular automata; self-replication; genetic evolution; diversity; adaptation S ince von Neumann's work on self-reproducing automata [1], artificial self-replication models based on cellular automata (CA) have formed one of the mainstreams in artificial life [2][3][4][5]. Recent developments indicate that simple CA with fixed rules can reproduce natural selection among different self-replicating structures [6,7], yet their evolutionary dynamics are considered quite limited in both diversity and adaptive behavior [8,9]. Contrary to these earlier observations, we show that genetic adaptation and diversification processes may occur in such simple CA. We investigate a system of evolving self-replicating loops, or evoloops [7], in which replication, variation and selection emerge solely from fixed local rules. Applying new tools for detailed genetic identification and genealogy tracing [10], we uncovered a genotypic permutation space that expands combinatorially with replicator size, within which loops exhibit significant diversity in macro-scale morphologies and mutational biases. Populations undergo nontrivial genetic adaptation, maximizing colony density while enhancing sustainability against other species. We also identified a set of nonmutable subsequences in the evoloop genetics, with which one can carry out genetic operations to alter fitness differentials and promote long-term evolutionary exploration. This effect is further manifested when populations are introduced to a hostile environment. These results demonstrate a unique example of genetic evolution that traverses multiple scales, hierarchically emerging from local interactions between elements much smaller than individual replicators.

Spatial-Pattern-Induced Evolution of a Self-Replicating Loop Network

“…In Sayama's earlier work, the size of loops was found to gradually decrease on average, and thus they cannot exhibit open-ended evolution. Further investigations showed that certain loop configurations can sustain and develop complexity of the loops [13,14]. We extend these studies by introducing a food-web structure between loops.…”

Self-Organization and Artificial Life

“…A CA consists of many units (cells), each of which can be in any of a number of discrete states, and each of which repeatedly determines its next state in a fully distributed manner, based on its current state and those of its neighbors. With no central controller involved, CAs can organize their state configurations to demonstrate various forms of selforganization: dynamical critical states such as in sand-pile models [15] and in the Game of Life [14], spontaneous formation of spatial patterns [47,211,216] (Figure 1(a)), self-replication 1 [116,117,158,178], and evolution by variation and natural selection [138,139,164,166,167,185]. Similarly, partial differential equations (PDEs), a continuous counterpart of CAs, have an even longer history of demonstrating self-organizing dynamics [52,76,142,190] (Figure 1(b)).…”

Evolvable Self-Reproducing Cells in a Two-Dimensional Artificial Chemistry

“…Self-replicating structures in CA range from von Neumann's marvelous but still impractical universal constructor [57,44] through to Langton's loops [24], and evoloops [47,46]. However, CA cannot typically be said to be material systems, since the state of each cell can usually be changed without cost.…”