A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities

• URL: http://arxiv.org/abs/2410.18938v1
• Date: Thu, 24 Oct 2024 17:24:34 GMT
• Title: A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities
• Authors: Yatin Dandi, Luca Pesce, Hugo Cui, Florent Krzakala, Yue M. Lu, Bruno Loureiro,
• Abstract summary: We study how fully-connected two-layer neural networks adapt to the target function after a single, but aggressive, gradient descent step. This provides a sharp description of the impact of feature learning in the generalization of two-layer neural networks, beyond the random features and lazy training regimes.
• Score: 30.737171081270322
• License:
• Abstract: A key property of neural networks is their capacity of adapting to data during training. Yet, our current mathematical understanding of feature learning and its relationship to generalization remain limited. In this work, we provide a random matrix analysis of how fully-connected two-layer neural networks adapt to the target function after a single, but aggressive, gradient descent step. We rigorously establish the equivalence between the updated features and an isotropic spiked random feature model, in the limit of large batch size. For the latter model, we derive a deterministic equivalent description of the feature empirical covariance matrix in terms of certain low-dimensional operators. This allows us to sharply characterize the impact of training in the asymptotic feature spectrum, and in particular, provides a theoretical grounding for how the tails of the feature spectrum modify with training. The deterministic equivalent further yields the exact asymptotic generalization error, shedding light on the mechanisms behind its improvement in the presence of feature learning. Our result goes beyond standard random matrix ensembles, and therefore we believe it is of independent technical interest. Different from previous work, our result holds in the challenging maximal learning rate regime, is fully rigorous and allows for finitely supported second layer initialization, which turns out to be crucial for studying the functional expressivity of the learned features. This provides a sharp description of the impact of feature learning in the generalization of two-layer neural networks, beyond the random features and lazy training regimes.

Related papers

• Analyzing Neural Scaling Laws in Two-Layer Networks with Power-Law Data Spectra [0.0]
  Neural scaling laws describe how the performance of deep neural networks scales with key factors such as training data size, model complexity, and training time. We employ techniques from statistical mechanics to analyze one-pass gradient descent within a student-teacher framework.
  arXiv  Detail & Related papers  (2024-10-11T17:21:42Z)
• Regularization, early-stopping and dreaming: a Hopfield-like setup to address generalization and overfitting [0.0]
  We look for optimal network parameters by applying a gradient descent over a regularized loss function. Within this framework, the optimal neuron-interaction matrices correspond to Hebbian kernels revised by a reiterated unlearning protocol.
  arXiv  Detail & Related papers  (2023-08-01T15:04:30Z)
• How Spurious Features Are Memorized: Precise Analysis for Random and NTK Features [19.261178173399784]
  We consider spurious features that are uncorrelated with the learning task. We provide a precise characterization of how they are memorized via two separate terms. We prove that the memorization of spurious features weakens as the generalization capability increases.
  arXiv  Detail & Related papers  (2023-05-20T05:27:41Z)
• Theoretical Characterization of the Generalization Performance of Overfitted Meta-Learning [70.52689048213398]
  This paper studies the performance of overfitted meta-learning under a linear regression model with Gaussian features. We find new and interesting properties that do not exist in single-task linear regression. Our analysis suggests that benign overfitting is more significant and easier to observe when the noise and the diversity/fluctuation of the ground truth of each training task are large.
  arXiv  Detail & Related papers  (2023-04-09T20:36:13Z)
• Modeling Uncertain Feature Representation for Domain Generalization [49.129544670700525]
  We show that our method consistently improves the network generalization ability on multiple vision tasks. Our methods are simple yet effective and can be readily integrated into networks without additional trainable parameters or loss constraints.
  arXiv  Detail & Related papers  (2023-01-16T14:25:02Z)
• Stabilizing Q-learning with Linear Architectures for Provably Efficient Learning [53.17258888552998]
  This work proposes an exploration variant of the basic $Q$-learning protocol with linear function approximation. We show that the performance of the algorithm degrades very gracefully under a novel and more permissive notion of approximation error.
  arXiv  Detail & Related papers  (2022-06-01T23:26:51Z)
• Analysis of Catastrophic Forgetting for Random Orthogonal Transformation Tasks in the Overparameterized Regime [9.184987303791292]
  We show that in permuted MNIST image classification tasks, the performance of multilayer perceptrons trained by vanilla gradient descent can be improved. We provide a theoretical explanation of this effect by studying a qualitatively similar two-task linear regression problem. We show that when a model is trained on the two tasks in sequence without any additional regularization, the risk gain on the first task is small.
  arXiv  Detail & Related papers  (2022-06-01T18:04:33Z)
• Going Beyond Linear RL: Sample Efficient Neural Function Approximation [76.57464214864756]
  We study function approximation with two-layer neural networks. Our results significantly improve upon what can be attained with linear (or eluder dimension) methods.
  arXiv  Detail & Related papers  (2021-07-14T03:03:56Z)
• Universal scaling laws in the gradient descent training of neural networks [10.508187462682308]
  We show that the learning trajectory can be characterized by an explicit bounds at large training times. Our results are based on spectral analysis of the evolution of a large network trained on the expected loss.
  arXiv  Detail & Related papers  (2021-05-02T16:46:38Z)
• A Functional Perspective on Learning Symmetric Functions with Neural Networks [48.80300074254758]
  We study the learning and representation of neural networks defined on measures. We establish approximation and generalization bounds under different choices of regularization. The resulting models can be learned efficiently and enjoy generalization guarantees that extend across input sizes.
  arXiv  Detail & Related papers  (2020-08-16T16:34:33Z)
• Can Temporal-Difference and Q-Learning Learn Representation? A Mean-Field Theory [110.99247009159726]
  Temporal-difference and Q-learning play a key role in deep reinforcement learning, where they are empowered by expressive nonlinear function approximators such as neural networks. In particular, temporal-difference learning converges when the function approximator is linear in a feature representation, which is fixed throughout learning, and possibly diverges otherwise.
  arXiv  Detail & Related papers  (2020-06-08T17:25:22Z)

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.