From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning

• URL: http://arxiv.org/abs/2502.03210v1
• Date: Wed, 05 Feb 2025 14:26:50 GMT
• Title: From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning
• Authors: Noa Rubin, Kirsten Fischer, Javed Lindner, David Dahmen, Inbar Seroussi, Zohar Ringel, Michael Krämer, Moritz Helias,
• Abstract summary: This work presents a theoretical framework of multi-scale adaptive feature learning bridging different approaches. A systematic expansion of the network's probability distribution reveals that mean-field scaling requires only a saddle-point approximation. Remarkably, we find across regimes that kernel adaptation can be reduced to an effective kernel rescaling when predicting the mean network output of a linear network.
• Score: 3.7857410821449755
• License:
• Abstract: Theoretically describing feature learning in neural networks is crucial for understanding their expressive power and inductive biases, motivating various approaches. Some approaches describe network behavior after training through a simple change in kernel scale from initialization, resulting in a generalization power comparable to a Gaussian process. Conversely, in other approaches training results in the adaptation of the kernel to the data, involving complex directional changes to the kernel. While these approaches capture different facets of network behavior, their relationship and respective strengths across scaling regimes remains an open question. This work presents a theoretical framework of multi-scale adaptive feature learning bridging these approaches. Using methods from statistical mechanics, we derive analytical expressions for network output statistics which are valid across scaling regimes and in the continuum between them. A systematic expansion of the network's probability distribution reveals that mean-field scaling requires only a saddle-point approximation, while standard scaling necessitates additional correction terms. Remarkably, we find across regimes that kernel adaptation can be reduced to an effective kernel rescaling when predicting the mean network output of a linear network. However, even in this case, the multi-scale adaptive approach captures directional feature learning effects, providing richer insights than what could be recovered from a rescaling of the kernel alone.

Related papers

• An Analytic Solution to Covariance Propagation in Neural Networks [10.013553984400488]
  This paper presents a sample-free moment propagation technique to accurately characterize the input-output distributions of neural networks. A key enabler of our technique is an analytic solution for the covariance of random variables passed through nonlinear activation functions. The wide applicability and merits of the proposed technique are shown in experiments analyzing the input-output distributions of trained neural networks and training Bayesian neural networks.
  arXiv  Detail & Related papers  (2024-03-24T14:08:24Z)
• A theory of data variability in Neural Network Bayesian inference [0.70224924046445]
  We provide a field-theoretic formalism which covers the generalization properties of infinitely wide networks. We derive the generalization properties from the statistical properties of the input. We show that data variability leads to a non-Gaussian action reminiscent of a ($varphi3+varphi4$)-theory.
  arXiv  Detail & Related papers  (2023-07-31T14:11:32Z)
• Simple initialization and parametrization of sinusoidal networks via their kernel bandwidth [92.25666446274188]
  sinusoidal neural networks with activations have been proposed as an alternative to networks with traditional activation functions. We first propose a simplified version of such sinusoidal neural networks, which allows both for easier practical implementation and simpler theoretical analysis. We then analyze the behavior of these networks from the neural tangent kernel perspective and demonstrate that their kernel approximates a low-pass filter with an adjustable bandwidth.
  arXiv  Detail & Related papers  (2022-11-26T07:41:48Z)
• Neural networks trained with SGD learn distributions of increasing complexity [78.30235086565388]
  We show that neural networks trained using gradient descent initially classify their inputs using lower-order input statistics. We then exploit higher-order statistics only later during training. We discuss the relation of DSB to other simplicity biases and consider its implications for the principle of universality in learning.
  arXiv  Detail & Related papers  (2022-11-21T15:27:22Z)
• Self-Consistent Dynamical Field Theory of Kernel Evolution in Wide Neural Networks [18.27510863075184]
  We analyze feature learning in infinite width neural networks trained with gradient flow through a self-consistent dynamical field theory. We construct a collection of deterministic dynamical order parameters which are inner-product kernels for hidden unit activations and gradients in each layer at pairs of time points.
  arXiv  Detail & Related papers  (2022-05-19T16:10:10Z)
• Inducing Gaussian Process Networks [80.40892394020797]
  We propose inducing Gaussian process networks (IGN), a simple framework for simultaneously learning the feature space as well as the inducing points. The inducing points, in particular, are learned directly in the feature space, enabling a seamless representation of complex structured domains. We report on experimental results for real-world data sets showing that IGNs provide significant advances over state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-04-21T05:27:09Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• The Principles of Deep Learning Theory [19.33681537640272]
  This book develops an effective theory approach to understanding deep neural networks of practical relevance. We explain how these effectively-deep networks learn nontrivial representations from training. We show that the depth-to-width ratio governs the effective model complexity of the ensemble of trained networks.
  arXiv  Detail & Related papers  (2021-06-18T15:00:00Z)
• Deep Archimedean Copulas [98.96141706464425]
  ACNet is a novel differentiable neural network architecture that enforces structural properties. We show that ACNet is able to both approximate common Archimedean Copulas and generate new copulas which may provide better fits to data.
  arXiv  Detail & Related papers  (2020-12-05T22:58:37Z)
• Learning Connectivity of Neural Networks from a Topological Perspective [80.35103711638548]
  We propose a topological perspective to represent a network into a complete graph for analysis. By assigning learnable parameters to the edges which reflect the magnitude of connections, the learning process can be performed in a differentiable manner. This learning process is compatible with existing networks and owns adaptability to larger search spaces and different tasks.
  arXiv  Detail & Related papers  (2020-08-19T04:53:31Z)
• Spectral Bias and Task-Model Alignment Explain Generalization in Kernel Regression and Infinitely Wide Neural Networks [17.188280334580195]
  Generalization beyond a training dataset is a main goal of machine learning. Recent observations in deep neural networks contradict conventional wisdom from classical statistics. We show that more data may impair generalization when noisy or not expressible by the kernel.
  arXiv  Detail & Related papers  (2020-06-23T17:53:11Z)

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.