Benign Overfitting in Single-Head Attention

• URL: http://arxiv.org/abs/2410.07746v1
• Date: Thu, 10 Oct 2024 09:23:33 GMT
• Title: Benign Overfitting in Single-Head Attention
• Authors: Roey Magen, Shuning Shang, Zhiwei Xu, Spencer Frei, Wei Hu, Gal Vardi,
• Abstract summary: We study benign overfitting in a single-head softmax attention model, which is the fundamental building block of Transformers. We prove that under appropriate conditions, the model exhibits benign overfitting in a classification setting already after two steps of gradient descent.
• Score: 27.297696573634976
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The phenomenon of benign overfitting, where a trained neural network perfectly fits noisy training data but still achieves near-optimal test performance, has been extensively studied in recent years for linear models and fully-connected/convolutional networks. In this work, we study benign overfitting in a single-head softmax attention model, which is the fundamental building block of Transformers. We prove that under appropriate conditions, the model exhibits benign overfitting in a classification setting already after two steps of gradient descent. Moreover, we show conditions where a minimum-norm/maximum-margin interpolator exhibits benign overfitting. We study how the overfitting behavior depends on the signal-to-noise ratio (SNR) of the data distribution, namely, the ratio between norms of signal and noise tokens, and prove that a sufficiently large SNR is both necessary and sufficient for benign overfitting.

Related papers

• Benign overfitting in leaky ReLU networks with moderate input dimension [23.522101525839687]
  We study benign overfitting in two-layer leaky ReLU networks trained with the hinge loss on a binary classification task. We characterize conditions on the signal to noise ratio (SNR) of the model parameters giving rise to benign versus non-benign (or harmful) overfitting.
  arXiv  Detail & Related papers  (2024-03-11T16:56:01Z)
• Learning with Noisy Foundation Models [95.50968225050012]
  This paper is the first work to comprehensively understand and analyze the nature of noise in pre-training datasets. We propose a tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization.
  arXiv  Detail & Related papers  (2024-03-11T16:22:41Z)
• Max-affine regression via first-order methods [7.12511675782289]
  The max-affine model ubiquitously arises in applications in signal processing and statistics. We present a non-asymptotic convergence analysis of gradient descent (GD) and mini-batch gradient descent (SGD) for max-affine regression.
  arXiv  Detail & Related papers  (2023-08-15T23:46:44Z)
• From Tempered to Benign Overfitting in ReLU Neural Networks [41.271773069796126]
  Overtrivialized neural networks (NNs) are observed to generalize well even when trained to perfectly fit noisy data. Recently, it was conjectured and empirically observed that the behavior of NNs is often better described as "tempered overfitting"
  arXiv  Detail & Related papers  (2023-05-24T13:36:06Z)
• Benign Overfitting in Linear Classifiers and Leaky ReLU Networks from KKT Conditions for Margin Maximization [59.038366742773164]
  Linears and leaky ReLU trained by gradient flow on logistic loss have an implicit bias towards satisfying the Karush-KuTucker (KKT) conditions. In this work we establish a number of settings where the satisfaction of these conditions implies benign overfitting in linear classifiers and in two-layer leaky ReLU networks.
  arXiv  Detail & Related papers  (2023-03-02T18:24:26Z)
• Benign Overfitting in Two-layer Convolutional Neural Networks [90.75603889605043]
  We study the benign overfitting phenomenon in training a two-layer convolutional neural network (CNN) We show that when the signal-to-noise ratio satisfies a certain condition, a two-layer CNN trained by gradient descent can achieve arbitrarily small training and test loss. On the other hand, when this condition does not hold, overfitting becomes harmful and the obtained CNN can only achieve constant level test loss.
  arXiv  Detail & Related papers  (2022-02-14T07:45:51Z)
• Benign Overfitting without Linearity: Neural Network Classifiers Trained by Gradient Descent for Noisy Linear Data [44.431266188350655]
  We consider the generalization error of two-layer neural networks trained to generalize by gradient descent. We show that neural networks exhibit benign overfitting: they can be driven to zero training error, perfectly fitting any noisy training labels, and simultaneously achieve minimax optimal test error. In contrast to previous work on benign overfitting that require linear or kernel-based predictors, our analysis holds in a setting where both the model and learning dynamics are fundamentally nonlinear.
  arXiv  Detail & Related papers  (2022-02-11T23:04:00Z)
• On the Double Descent of Random Features Models Trained with SGD [78.0918823643911]
  We study properties of random features (RF) regression in high dimensions optimized by gradient descent (SGD) We derive precise non-asymptotic error bounds of RF regression under both constant and adaptive step-size SGD setting. We observe the double descent phenomenon both theoretically and empirically.
  arXiv  Detail & Related papers  (2021-10-13T17:47:39Z)
• The Interplay Between Implicit Bias and Benign Overfitting in Two-Layer Linear Networks [51.1848572349154]
  neural network models that perfectly fit noisy data can generalize well to unseen test data. We consider interpolating two-layer linear neural networks trained with gradient flow on the squared loss and derive bounds on the excess risk.
  arXiv  Detail & Related papers  (2021-08-25T22:01:01Z)
• Provable Model-based Nonlinear Bandit and Reinforcement Learning: Shelve Optimism, Embrace Virtual Curvature [61.22680308681648]
  We show that global convergence is statistically intractable even for one-layer neural net bandit with a deterministic reward. For both nonlinear bandit and RL, the paper presents a model-based algorithm, Virtual Ascent with Online Model Learner (ViOL)
  arXiv  Detail & Related papers  (2021-02-08T12:41:56Z)
• Binary Classification of Gaussian Mixtures: Abundance of Support Vectors, Benign Overfitting and Regularization [39.35822033674126]
  We study binary linear classification under a generative Gaussian mixture model. We derive novel non-asymptotic bounds on the classification error of the latter. Our results extend to a noisy model with constant probability noise flips.
  arXiv  Detail & Related papers  (2020-11-18T07:59:55Z)

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.