Abstract:For the goal of automatically evolving Embodied Intelligence (EI), we investigate an open software architecture inspired by the high surface area to volume ratio of animal lungs, which aims to avoid information theoretic limits on long term evolution experiments (LTEE) encountered with monolithic genetic programming trees. Instead individuals are teams composed of 1023 trees whose inputs and outputs are linked by a low entropy loss branching data (air) pathway. Most trees are shallow and software engineering's failed disruption propagation (FDP) is observed in the small fraction of deep trees. After initial search, most improvements are at intermediate depths and performance is still rising even after 100 000 generations. Despite the use of double precision for the bifurcating data interconnect, some information loss is seen, particularly in early generations. The static optimisation benchmark, appears to encourage early convergence, which locks the population into possibly sub-optimal phenotypes. Later thousands of small improvements, sometimes in large bloated ensemble members, appear to compensate for early overfitting. Using tournament fitness selection and subtree crossover, we target pure nested side-effect free floating point functions, which are known to have low FDP, and high fidelity data paths, in the hope of generating code which is not too robust so as to prevent on going improvement. However, we again find genetic changes deep within trees are silent. For single precision, we find a maximum evolvability sweet spot with trees of depth 10 to 100. Accordingly, we suggest to evolve very large very complex programs needed for Embodied Intelligence, an open structure with a high surface area permitting most mutation sites to be within 10-100 levels of the organism's environment, and many better placed test oracles to monitor the impact of mutated code, will be needed.

Evolutionary Learning in the 2D Artificial Life System "Avida"

Learning and Complexity in Genetic Auto-Adaptive Systems

Towards Large-Scale Simulations of Open-Ended Evolution in Continuous Cellular Automata

Toward Artificial Open-Ended Evolution within Lenia using Quality-Diversity

Survival and Evolutionary Adaptation of Populations Under Disruptive Habitat Change: A Study With Darwinian Cellular Automata

Adaptability and Homeostasis in the Game of Life interacting with the evolved Cellular Automata

Artificial Life using a Book and Bookmarker

Open to Evolve Embodied Intelligence

Emergent Dynamics in Heterogeneous Life-Like Cellular Automata

Transfer Dynamics in Emergent Evolutionary Curricula

Evolution-based Shape and Behavior Co-design of Virtual Agents

An evolutionary game with environmental feedback and players' opinions

Meta-Learning an Evolvable Developmental Encoding

Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games

Synergetic Learning Systems: Concept, Architecture, and Algorithms

Evolution and learning in differentiable robots

Flow-Lenia: Towards open-ended evolution in cellular automata through mass conservation and parameter localization

Self-Reproduction and Evolution in Cellular Automata: 25 Years after Evoloops

Evolution with Opponent-Learning Awareness

Imitation of Life: Advanced system for native Artificial Evolution

Mutation-Bias Learning in Games