Evolution of Community Complexity Workshop

10:00-10:45 (10:00-10:30 Talk, 10:30-10:45 Discussion)

Seth Bullock (The University of Bristol)

Evolving bi-directional referential communication in minimal autonomous agents

Referential communication is central to social and collective behaviour, e.g., honey bees communicating nectar locations to each other or co-workers gossiping about a colleague. Typically, such behaviour is considered to be “representation hungry”, requiring complex cognitive machinery capable of manipulating symbolic representations of the world. However, a series of simulation studies have shown that it can be achieved by very simple embodied artificial agents controlled by small evolved continuous time recurrent neural networks (CTRNNs) that are challenging to interpret in symbol processing terms.

10:45-11:30 (10:45-11:15 Talk, 11:15-11:30 Discussion)

Mizuki Oka (University of Tsukuba)

Towards Open-Ended Evolution in Artificial Systems

Open-Ended Evolution (OEE), a significant subfield of Artificial Life (ALife), focuses on realizing sustained, unbounded evolution within artificial systems. This presentation explores various mechanisms and algorithms to facilitate open-ended evolution, employing agent-based models and simulations of virtual organisms as research tools.

14:30-15:15 (14:30-15:00 Talk, 15:00-15:15 Discussion)

Acer Chang (Rikkyo University)

Between Individual Brains and Collective Behavior: Multi-level Emergence in a Group Formation Task

Emergence is a property often claimed to apply to complex systems on multiple levels of organization: individual behavior emerges from underlying neural activity, social patterns — from constituent behaviors of the individuals. Furthermore, emergent level is typically characterized as possessing autonomy from the lower-level phenomena and as exerting downward causation on them. In this study we investigate such a multi-level emergence in the context of a single simple task. We evolve agents controlled by a small neural network to travel in formation. We then compute measures of emergence stemming from an approach known as Integrated Information Decomposition. Results are presented for both final behavior and the evolutionary changes that led to it.

15:15-16:00 (15:15-15:45 Talk, 15:45-16:00 Discussion)

Takashi Ikegami (The University of Tokyo)

Evolution of Individuality

The definition of "individuality" in life systems is not self-evident. An individual is considered a unit of evolution, but what defines an individual is not only a physical boundary but also an informational boundary. It may be something that extends temporally rather than spatially. It is the agency of an individual, its memory, and also the result of group evolution. In this presentation, I will consider the evolution of individuality, the evolution of autonomy and diversity, based on the hypothesis that "they are born within a group", based on specific phenomena.