When to be critical? Performance and evolvability in different regimes of neural Ising agents

• URL: http://arxiv.org/abs/2303.16195v4
• Date: Fri, 24 Nov 2023 16:02:00 GMT
• Title: When to be critical? Performance and evolvability in different regimes of neural Ising agents
• Authors: Sina Khajehabdollahi, Jan Prosi, Emmanouil Giannakakis, Georg Martius, Anna Levina
• Abstract summary: 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. Surprisingly, we find that all populations that discover solutions, evolve to be subcritical.
• Score: 18.536813548129878
• License: http://creativecommons.org/licenses/by-sa/4.0/
• 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.

Related papers

• Cognitive Evolutionary Learning to Select Feature Interactions for Recommender Systems [59.117526206317116]
  We show that CELL can adaptively evolve into different models for different tasks and data. Experiments on four real-world datasets demonstrate that CELL significantly outperforms state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-05-29T02:35:23Z)
• Agent Alignment in Evolving Social Norms [65.45423591744434]
  We propose an evolutionary framework for agent evolution and alignment, named EvolutionaryAgent. In an environment where social norms continuously evolve, agents better adapted to the current social norms will have a higher probability of survival and proliferation. We show that EvolutionaryAgent can align progressively better with the evolving social norms while maintaining its proficiency in general tasks.
  arXiv  Detail & Related papers  (2024-01-09T15:44:44Z)
• Causal machine learning for single-cell genomics [94.28105176231739]
  We discuss the application of machine learning techniques to single-cell genomics and their challenges. We first present the model that underlies most of current causal approaches to single-cell biology. We then identify open problems in the application of causal approaches to single-cell data.
  arXiv  Detail & Related papers  (2023-10-23T13:35:24Z)
• Phylogeny-informed fitness estimation [58.720142291102135]
  We propose phylogeny-informed fitness estimation, which exploits a population's phylogeny to estimate fitness evaluations. Our results indicate that phylogeny-informed fitness estimation can mitigate the drawbacks of down-sampled lexicase. This work serves as an initial step toward improving evolutionary algorithms by exploiting runtime phylogenetic analysis.
  arXiv  Detail & Related papers  (2023-06-06T19:05:01Z)
• The scaling of goals via homeostasis: an evolutionary simulation, experiment and analysis [0.0]
  We propose that evolution pivoted the collective intelligence of cells during morphogenesis into behavioral intelligence by scaling up the goal states at the center of homeostatic processes. We found that these emergent morphogenetic agents exhibit a number of predicted features, including the use of stress propagation dynamics to achieve its target morphology. We propose that this system is a first step toward a quantitative understanding of how evolution scales minimal goal-directed behavior (homeostatic loops) into higher-level problem-solving agents in morphogenetic and other spaces.
  arXiv  Detail & Related papers  (2022-11-15T21:48:44Z)
• Towards the Neuroevolution of Low-level Artificial General Intelligence [5.2611228017034435]
  We argue that the search for Artificial General Intelligence (AGI) should start from a much lower level than human-level intelligence. Our hypothesis is that learning occurs through sensory feedback when an agent acts in an environment. We evaluate a method to evolve a biologically-inspired artificial neural network that learns from environment reactions.
  arXiv  Detail & Related papers  (2022-07-27T15:30:50Z)
• Transfer Dynamics in Emergent Evolutionary Curricula [4.692078300163222]
  PINSKY is a system for open-ended learning through neuroevolution in game-based domains. This paper focuses on the role of transfer of policies from one evolutionary branch ("species") to another. The most insightful finding is that inter-species transfer, while rare, is crucial to the system's success.
  arXiv  Detail & Related papers  (2022-03-03T21:10:22Z)
• The Introspective Agent: Interdependence of Strategy, Physiology, and Sensing for Embodied Agents [51.94554095091305]
  We argue for an introspective agent, which considers its own abilities in the context of its environment. Just as in nature, we hope to reframe strategy as one tool, among many, to succeed in an environment.
  arXiv  Detail & Related papers  (2022-01-02T20:14:01Z)
• The dynamical regime and its importance for evolvability, task performance and generalization [14.059479351946386]
  We find that all populations, regardless of their initial regime, evolve to be subcritical in simple tasks. We conclude that although the subcritical regime is preferable for a simple task, the optimal deviation from criticality depends on the task difficulty.
  arXiv  Detail & Related papers  (2021-03-22T21:22:52Z)
• Task-Agnostic Morphology Evolution [94.97384298872286]
  Current approaches that co-adapt morphology and behavior use a specific task's reward as a signal for morphology optimization. This often requires expensive policy optimization and results in task-dependent morphologies that are not built to generalize. We propose a new approach, Task-Agnostic Morphology Evolution (TAME), to alleviate both of these issues.
  arXiv  Detail & Related papers  (2021-02-25T18:59:21Z)
• Self-adaptation in non-Elitist Evolutionary Algorithms on Discrete Problems with Unknown Structure [2.28438857884398]
  Key challenge to make effective use of evolutionary algorithms is to choose appropriate settings for their parameters. Non-deterministic parameter control mechanisms adjust parameters using information obtained from the evolutionary process. Self-adaptation is a popular parameter control mechanism in evolutionary strategies.
  arXiv  Detail & Related papers  (2020-04-01T10:35:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.