Simulating Evolvability as a Learning Algorithm: Empirical Investigations on Distribution Sensitivity, Robustness, and Constraint Tradeoffs

• URL: http://arxiv.org/abs/2507.18666v1
• Date: Thu, 24 Jul 2025 04:32:31 GMT
• Title: Simulating Evolvability as a Learning Algorithm: Empirical Investigations on Distribution Sensitivity, Robustness, and Constraint Tradeoffs
• Authors: Nicholas Fidalgo, Puyuan Ye,
• Abstract summary: Theory of evolvability formalizes evolution as a constrained learning algorithm operating without labeled examples or structural knowledge.<n>We implement a genetic algorithm that faithfully simulates Valiant's model and conduct experiments across six Boolean function classes.<n>Our findings reveal sharp performance drops at intermediate dimensions and expose the essential role of neutral mutations in escaping fitness plateaus.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The theory of evolvability, introduced by Valiant (2009), formalizes evolution as a constrained learning algorithm operating without labeled examples or structural knowledge. While theoretical work has established the evolvability of specific function classes under idealized conditions, the framework remains largely untested empirically. In this paper, we implement a genetic algorithm that faithfully simulates Valiant's model and conduct extensive experiments across six Boolean function classes: monotone conjunctions, monotone disjunctions, parity, majority, general conjunctions, and general disjunctions. Our study examines evolvability under uniform and non-uniform distributions, investigates the effects of fixed initial hypotheses and the removal of neutral mutations, and highlights how these constraints alter convergence behavior. We validate known results (e.g., evolvability of monotone conjunctions, non-evolvability of parity) and offer the first empirical evidence on the evolvability of majority and general Boolean classes. Our findings reveal sharp performance drops at intermediate dimensions and expose the essential role of neutral mutations in escaping fitness plateaus. We also demonstrate that evolvability can depend strongly on the input distribution. These insights clarify practical limits of evolutionary search and suggest new directions for theoretical work, including potential refinements to evolvability definitions and bounds. Our implementation provides a rigorous, extensible framework for empirical analysis and serves as a testbed for future explorations of learning through evolution.

Related papers

• Towards Understanding Extrapolation: a Causal Lens [53.15488984371969]
  We provide a theoretical understanding of when extrapolation is possible and offer principled methods to achieve it.<n>Under this formulation, we cast the extrapolation problem into a latent-variable identification problem.<n>Our theory reveals the intricate interplay between the underlying manifold's smoothness and the shift properties.
  arXiv  Detail & Related papers  (2025-01-15T21:29:29Z)
• Causal Representation Learning from Multimodal Biomedical Observations [57.00712157758845]
  We develop flexible identification conditions for multimodal data and principled methods to facilitate the understanding of biomedical datasets.<n>Key theoretical contribution is the structural sparsity of causal connections between modalities.<n>Results on a real-world human phenotype dataset are consistent with established biomedical research.
  arXiv  Detail & Related papers  (2024-11-10T16:40:27Z)
• Toward Understanding In-context vs. In-weight Learning [50.24035812301655]
  We identify simplified distributional properties that give rise to the emergence and disappearance of in-context learning.<n>We then extend the study to a full large language model, showing how fine-tuning on various collections of natural language prompts can elicit similar in-context and in-weight learning behaviour.
  arXiv  Detail & Related papers  (2024-10-30T14:09:00Z)
• Cross-Entropy Is All You Need To Invert the Data Generating Process [29.94396019742267]
  Empirical phenomena suggest that supervised models can learn interpretable factors of variation in a linear fashion.<n>Recent advances in self-supervised learning have shown that these methods can recover latent structures by inverting the data generating process.<n>We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation.
  arXiv  Detail & Related papers  (2024-10-29T09:03:57Z)
• Range, not Independence, Drives Modularity in Biologically Inspired Representations [52.48094670415497]
  We develop a theory of when biologically inspired networks modularise their representation of source variables (sources)<n>We derive necessary and sufficient conditions on a sample of sources that determine whether the neurons in an optimal linear autoencoder modularise.<n>Our theory applies to any dataset, extending far beyond the case of statistical independence studied in previous work.
  arXiv  Detail & Related papers  (2024-10-08T17:41:37Z)
• Geometric Understanding of Discriminability and Transferability for Visual Domain Adaptation [27.326817457760725]
  Invariant representation learning for unsupervised domain adaptation (UDA) has made significant advances in computer vision and pattern recognition communities. Recently, empirical connections between transferability and discriminability have received increasing attention. In this work, we systematically analyze the essentials of transferability and discriminability from the geometric perspective.
  arXiv  Detail & Related papers  (2024-06-24T13:31:08Z)
• Class-wise Activation Unravelling the Engima of Deep Double Descent [0.0]
  Double descent presents a counter-intuitive aspect within the machine learning domain. In this study, we revisited the phenomenon of double descent and discussed the conditions of its occurrence.
  arXiv  Detail & Related papers  (2024-05-13T12:07:48Z)
• Identifiable Latent Neural Causal Models [82.14087963690561]
  Causal representation learning seeks to uncover latent, high-level causal representations from low-level observed data. We determine the types of distribution shifts that do contribute to the identifiability of causal representations. We translate our findings into a practical algorithm, allowing for the acquisition of reliable latent causal representations.
  arXiv  Detail & Related papers  (2024-03-23T04:13:55Z)
• Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning [80.44084021062105]
  We propose a novel latent partial causal model for multimodal data, featuring two latent coupled variables, connected by an undirected edge, to represent the transfer of knowledge across modalities.<n>Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by multimodal contrastive learning correspond to the latent coupled variables up to a trivial transformation.<n>Experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets.
  arXiv  Detail & Related papers  (2024-02-09T07:18:06Z)
• On the Joint Interaction of Models, Data, and Features [82.60073661644435]
  We introduce a new tool, the interaction tensor, for empirically analyzing the interaction between data and model through features. Based on these observations, we propose a conceptual framework for feature learning. Under this framework, the expected accuracy for a single hypothesis and agreement for a pair of hypotheses can both be derived in closed-form.
  arXiv  Detail & Related papers  (2023-06-07T21:35:26Z)
• Multi-Study Boosting: Theoretical Considerations for Merging vs. Ensembling [2.252304836689618]
  Cross-study replicability is a powerful model evaluation criterion that emphasizes generalizability of predictions. We study boosting algorithms in the presence of potential heterogeneity in predictor-outcome relationships across studies. We compare two multi-study learning strategies: 1) merging all the studies and training a single model, and 2) multi-study ensembling.
  arXiv  Detail & Related papers  (2022-07-11T02:25:47Z)

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.