An Overview of Open-Ended Evolution: Editorial Introduction to the Open-Ended Evolution II Special Issue

Open-Endedness in Genelife

We explore the open-ended nature of evolution in Genelife, an evolutionary extension of Conway's Game of Life cellular automaton in which "live" cell states are endowed at birth with a genome that affects their local dynamics and can be inherited. Both genetic sequences and locally connected spatial patterns are analyzed for novelty, keeping track of all new structures, and innovation is quantified using activity statistics. The impacts of both spatial symmetry breaking with nontotalistic rules and superimposed density regulation of the live state proliferation on the open-ended nature of the evolution are explored. Conditions are found where both genetic and spatial patterns exhibit open-ended innovation. This innovation appears to fall short of functional biological innovation, however, and potential reasons for this are discussed.

On the Open-Endedness of Detecting Open-Endedness

We argue that attempting to quantify open-endedness misses the point: The nature of open-endedness is such that an open-ended system will eventually move outside its current model of behavior, and hence outside any measure based on that model. This presents a challenge for analyzing Artificial Life systems, leading us to conclude that the focus should be on understanding the mechanisms underlying open-endedness, not simply on attempting to quantify it. To demonstrate this, we apply several measures to eight long experimental runs of the spatial version of the Stringmol automata chemistry. These experiments were originally designed to examine the hypothesis that spatial structure provides a defense against parasites. The runs successfully show this defense, but also show a range of innovative, and possibly open-ended, behaviors involved in countering a parasitic arms race. Commencing with system-generic measures, we develop and use a variety of measures dedicated to analyzing some of these innovations. We argue that a process of analysis, starting with system-generic measures but going on to system-specific measures, will be needed wherever the phenomenon of open-endedness is involved.

Kuhnian Lessons for the Study of Open-Ended Evolution

Kuhnian philosophy of science implies that progress in the study of open-ended evolution (OEE) would be accelerated if the OEE science community were to agree on some examples of striking success in OEE science. This article recounts the important role of scientific paradigms and scientific exemplars in creating the productivity of what Kuhn, in The Structure of Scientific Revolutions, calls "normal" science, and it describes how the study of OEE today would benefit from exhibiting more of the hallmarks of normal science. The article concludes by describing five proposed projects that would help create a consensus in the OEE community on some good examples of the scientific study of OEE.

A Procedure for Testing for Tokyo Type 1 Open-Ended Evolution

Tokyo Type 1 open-ended evolution (OEE) is a category of OEE that includes systems exhibiting the ongoing generation of adaptive novelty and ongoing growth in complexity. It can be considered as a necessary foundation for Tokyo Type 2 OEE (ongoing evolution of evolvability) and Tokyo Type 3 OEE (ongoing generation of major transitions). This article brings together five methods of analysis to form a procedure for testing for Tokyo Type 1 OEE. The procedure is presented as simply as possible, isolated from the complexities of any particular evolutionary system, and with a clear rationale for each step. In developing these steps, we also identify five key challenges in OEE. The last of these (achieving a higher order of complexity growth within a system exhibiting indefinitely scalable complexity) can be considered a grand challenge for Tokyo Type 1 OEE. Promising approaches to this grand challenge include also achieving one or both of Tokyo Types 2 and 3 OEE; this can be seen as one answer to why these other types of OEE are important, providing a unified view of OEE.

Open-ended versus bounded evolution: Mineral evolution as a case study

To what extent are naturally evolving systems limited in their potential diversity (i.e. "bounded") versus unrestricted ("open-ended")? Minerals provide a quantitative model evolving system, with well-documented increases in mineral diversity through multiple stages of planetary evolution over billions of years. A recent framework that unifies behaviors of both biotic and abiotic evolving systems posits that all such systems are characterized by combinatorial richness subject to selection. In the case of minerals, combinatorial richness derives from the possible combinations of chemical elements coupled with permutations of their formulas' coefficients. Observed mineral species, which are selected for persistence through deep time, represent a miniscule fraction of all possible element configurations. Furthermore, this model predicts that as planetary systems evolve, stable minerals become an ever-smaller fraction of the "possibility space." A postulate is that "functional information," defined as the negative log2 of that fraction, must increase as a system evolves. We have tested this hypothesis for minerals by estimating the fraction of all possible chemical formulas observed from one stage of mineral evolution to the next, based on numbers of different essential elements and the maximum chemical formula complexity at each of nine chronological stages of mineral evolution. We find a monotonic increase in mineral functional information through these nine stages-a result consistent with the hypothesis. Furthermore, analysis of the chemical formulas of minerals demonstrates that the modern Earth may be approaching the maximum limit of functional information for natural mineral systems-a result demonstrating that mineral evolution is not open-ended.

Chapter 2: What Is Life?

The question "What is life?" has existed since the beginning of recorded history. However, the scientific and philosophical contexts of this question have changed and been refined as advancements in technology have revealed both fine details and broad connections in the network of life on Earth. Understanding the framework of the question "What is life?" is central to formulating other questions such as "Where else could life be?" and "How do we search for life elsewhere?" While many of these questions are addressed throughout the Astrobiology Primer 3.0, this chapter gives historical context for defining life, highlights conceptual characteristics shared by all life on Earth as well as key features used to describe it, discusses why it matters for astrobiology, and explores both challenges and opportunities for finding an informative operational definition.

Open-endedness in synthetic biology: A route to continual innovation for biological design

Design in synthetic biology is typically goal oriented, aiming to repurpose or optimize existing biological functions, augmenting biology with new-to-nature capabilities, or creating life-like systems from scratch. While the field has seen many advances, bottlenecks in the complexity of the systems built are emerging and designs that function in the lab often fail when used in real-world contexts. Here, we propose an open-ended approach to biological design, with the novelty of designed biology being at least as important as how well it fulfils its goal. Rather than solely focusing on optimization toward a single best design, designing with novelty in mind may allow us to move beyond the diminishing returns we see in performance for most engineered biology. Research from the artificial life community has demonstrated that embracing novelty can automatically generate innovative and unexpected solutions to challenging problems beyond local optima. Synthetic biology offers the ideal playground to explore more creative approaches to biological design.

Does the Field of Nature-Inspired Computing Contribute to Achieving Lifelike Features?

The main idea behind artificial intelligence was simple: what if we study living systems to develop new, practical computing systems that possess "lifelike" properties? And that's exactly how evolutionary computing emerged. Researchers came up with ideas inspired by the principles of evolution to develop intelligent methods to tackle hard problems. The efficacy of these methods made researchers seek inspiration in living organisms and systems and extend the evolutionary concept to other nature-inspired ideas. In recent years, nature-inspired computing has exhibited an exponential increase in the number of algorithms that are presented each year. Authors claim that they are inspired by a behavior found in nature to come up with a lifelike algorithm. However, the mathematical background does not match the behavior in the majority of these cases. Thus the question is, do all nature-inspired algorithms remain lifelike? Also, are there any ideas included that contribute to computing? This study aims to (a) present some nature-inspired methods that contribute to achieving lifelike features of computing systems and (b) discuss if there is any need for new lifelike features.

DigiHive: Artificial Chemistry Environment for Modeling of Self-Organization Phenomena

The article presents the DigiHive system, an artificial chemistry simulation environment, and the results of preliminary simulation experiments leading toward building a self-replicating system resembling a living cell. The two-dimensional environment is populated by particles that can bond together and form complexes of particles. Some complexes can recognize and change the structures of surrounding complexes, where the functions they perform are encoded in their structure in the form of Prolog-like language expressions. After introducing the DigiHive environment, we present the results of simulations of two fundamental parts of a self-replicating system, the work of a universal constructor and a copying machine, and the growth and division of a cell-like wall. At the end of the article, the limitations and arising difficulties of modeling in the DigiHive environment are presented, along with a discussion of possible future experiments and applications of this type of modeling.

The Typical Set and Entropy in Stochastic Systems with Arbitrary Phase Space Growth

The existence of the typical set is key for data compression strategies and for the emergence of robust statistical observables in macroscopic physical systems. Standard approaches derive its existence from a restricted set of dynamical constraints. However, given its central role underlying the emergence of stable, almost deterministic statistical patterns, a question arises whether typical sets exist in much more general scenarios. We demonstrate here that the typical set can be defined and characterized from general forms of entropy for a much wider class of stochastic processes than was previously thought. This includes processes showing arbitrary path dependence, long range correlations or dynamic sampling spaces, suggesting that typicality is a generic property of stochastic processes, regardless of their complexity. We argue that the potential emergence of robust properties in complex stochastic systems provided by the existence of typical sets has special relevance to biological systems.

Recent trends in robot learning and evolution for swarm robotics

Swarm robotics is a promising approach to control large groups of robots. However, designing the individual behavior of the robots so that a desired collective behavior emerges is still a major challenge. In recent years, many advances in the automatic design of control software for robot swarms have been made, thus making automatic design a promising tool to address this challenge. In this article, I highlight and discuss recent advances and trends in offline robot evolution, embodied evolution, and offline robot learning for swarm robotics. For each approach, I describe recent design methods of interest, and commonly encountered challenges. In addition to the review, I provide a perspective on recent trends and discuss how they might influence future research to help address the remaining challenges of designing robot swarms.

Evolving neural networks through bio-inspired parent selection in dynamic environments

Environmental variability often degrades the performance of algorithms designed to capture the global convergence of a given search space. Several approaches have been developed to challenge environmental uncertainty by incorporating biologically inspired notions, focusing on crossover, mutation, and selection. This study proposes a bio-inspired approach called NEAT-HD, which focuses on parent selection based on genetic similarity. The originality of the proposed approach rests on its use of a sigmoid function to accelerate species formation and contribute to population diversity. Experiments on two classic control tasks were performed to demonstrate the performance of the proposed method. The results show that NEAT-HD can dynamically adapt to its environment by forming hybrid individuals originating from genetically distinct parents. Additionally, an increase in diversity within the population was observed due to the formation of hybrids and novel individuals, which has never been observed before. Comparing two tasks, the characteristics of NEAT-HD were improved by appropriately setting the algorithm to include the distribution of genetic distance within the population. Our key finding is the inherent potential of newly formed individuals for robustness against dynamic environments.

Self-replication of a quantum artificial organism driven by single-photon pulses

Imitating the transition from inanimate to living matter is a longstanding challenge. Artificial life has achieved computer programs that self-replicate, mutate, compete and evolve, but lacks self-organized hardwares akin to the self-assembly of the first living cells. Nonequilibrium thermodynamics has achieved lifelike self-organization in diverse physical systems, but has not yet met the open-ended evolution of living organisms. Here, I look for the emergence of an artificial-life code in a nonequilibrium physical system undergoing self-organization. I devise a toy model where the onset of self-replication of a quantum artificial organism (a chain of lambda systems) is owing to single-photon pulses added to a zero-temperature environment. I find that spontaneous mutations during self-replication are unavoidable in this model, due to rare but finite absorption of off-resonant photons. I also show that the replication probability is proportional to the absorbed work from the photon, thereby fulfilling a dissipative adaptation (a thermodynamic mechanism underlying lifelike self-organization). These results hint at self-replication as the scenario where dissipative adaptation (pointing towards convergence) coexists with open-ended evolution (pointing towards divergence).