When to be critical? Performance and evolvability in different regimes of neural Ising agents
arXiv:2303.16195 · doi:10.1162/artl_a_00383
Abstract
It has long been hypothesized that operating close to the critical state is beneficial for natural, artificial and their evolutionary systems. We put this hypothesis to test in a system of evolving foraging agents controlled by neural networks that can adapt agents' dynamical regime throughout evolution. Surprisingly, we find that all populations that discover solutions, evolve to be subcritical. By a resilience analysis, we find that there are still benefits of starting the evolution in the critical regime. Namely, initially critical agents maintain their fitness level under environmental changes (for example, in the lifespan) and degrade gracefully when their genome is perturbed. At the same time, initially subcritical agents, even when evolved to the same fitness, are often inadequate to withstand the changes in the lifespan and degrade catastrophically with genetic perturbations. Furthermore, we find the optimal distance to criticality depends on the task complexity. To test it we introduce a hard and simple task: for the hard task, agents evolve closer to criticality whereas more subcritical solutions are found for the simple task. We verify that our results are independent of the selected evolutionary mechanisms by testing them on two principally different approaches: a genetic algorithm and an evolutionary strategy. In summary, our study suggests that although optimal behaviour in the simple task is obtained in a subcritical regime, initializing near criticality is important to be efficient at finding optimal solutions for new tasks of unknown complexity.
arXiv commentarXiv admin note: substantial text overlap with arXiv:2103.12184
References in corpus (16)
- Weak pairwise correlations imply strongly correlated network states in a neural population
- Scale-free correlations in bird flocks
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- Optimal Dynamical Range of Excitable Networks at Criticality
- Are biological systems poised at criticality?
- Colloquium: Criticality and dynamical scaling in living systems
- Information-based fitness and the emergence of criticality in living systems
- Control of criticality and computation in spiking neuromorphic networks with plasticity
- Between perfectly critical and fully irregular: a reverberating model captures and predicts cortical spike propagation
- Self-organization toward criticality by synaptic plasticity
- Feedback mechanisms for self-organization to the edge of a phase transition
- A Critical-like Collective State Leads to Long-range Cell Communication in Dictyostelium discoideum Aggregation
- Evolving generalists via dynamic sculpting of rugged landscapes
- Tailored ensembles of neural networks optimize sensitivity to stimulus statistics
- Intrinsic noise and deviations from criticality in Boolean gene-regulatory networks
- Criticality as It Could Be: organizational invariance as self-organized criticality in embodied agents