Linear Stability Hypothesis and Rank Stratification for Nonlinear Models

• URL: http://arxiv.org/abs/2211.11623v1
• Date: Mon, 21 Nov 2022 16:27:25 GMT
• Title: Linear Stability Hypothesis and Rank Stratification for Nonlinear Models
• Authors: Yaoyu Zhang, Zhongwang Zhang, Leyang Zhang, Zhiwei Bai, Tao Luo, Zhi-Qin John Xu
• Abstract summary: We propose a rank stratification for general nonlinear models to uncover a model rank as an "effective size of parameters" By these results, model rank of a target function predicts a minimal training data size for its successful recovery.
• Score: 3.0041514772139166
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Models with nonlinear architectures/parameterizations such as deep neural networks (DNNs) are well known for their mysteriously good generalization performance at overparameterization. In this work, we tackle this mystery from a novel perspective focusing on the transition of the target recovery/fitting accuracy as a function of the training data size. We propose a rank stratification for general nonlinear models to uncover a model rank as an "effective size of parameters" for each function in the function space of the corresponding model. Moreover, we establish a linear stability theory proving that a target function almost surely becomes linearly stable when the training data size equals its model rank. Supported by our experiments, we propose a linear stability hypothesis that linearly stable functions are preferred by nonlinear training. By these results, model rank of a target function predicts a minimal training data size for its successful recovery. Specifically for the matrix factorization model and DNNs of fully-connected or convolutional architectures, our rank stratification shows that the model rank for specific target functions can be much lower than the size of model parameters. This result predicts the target recovery capability even at heavy overparameterization for these nonlinear models as demonstrated quantitatively by our experiments. Overall, our work provides a unified framework with quantitative prediction power to understand the mysterious target recovery behavior at overparameterization for general nonlinear models.

Related papers

• Scaling and renormalization in high-dimensional regression [72.59731158970894]
  This paper presents a succinct derivation of the training and generalization performance of a variety of high-dimensional ridge regression models. We provide an introduction and review of recent results on these topics, aimed at readers with backgrounds in physics and deep learning.
  arXiv  Detail & Related papers  (2024-05-01T15:59:00Z)
• Bayesian Inference for Consistent Predictions in Overparameterized Nonlinear Regression [0.0]
  This study explores the predictive properties of over parameterized nonlinear regression within the Bayesian framework. Posterior contraction is established for generalized linear and single-neuron models with Lipschitz continuous activation functions. The proposed method was validated via numerical simulations and a real data application.
  arXiv  Detail & Related papers  (2024-04-06T04:22:48Z)
• Optimal Nonlinearities Improve Generalization Performance of Random Features [0.9790236766474201]
  Random feature model with a nonlinear activation function has been shown to performally equivalent to a Gaussian model in terms of training and generalization errors. We show that acquired parameters from the Gaussian model enable us to define a set of optimal nonlinearities. Our numerical results validate that the optimized nonlinearities achieve better generalization performance than widely-used nonlinear functions such as ReLU.
  arXiv  Detail & Related papers  (2023-09-28T20:55:21Z)
• Kalman Filter for Online Classification of Non-Stationary Data [101.26838049872651]
  In Online Continual Learning (OCL) a learning system receives a stream of data and sequentially performs prediction and training steps. We introduce a probabilistic Bayesian online learning model by using a neural representation and a state space model over the linear predictor weights. In experiments in multi-class classification we demonstrate the predictive ability of the model and its flexibility to capture non-stationarity.
  arXiv  Detail & Related papers  (2023-06-14T11:41:42Z)
• 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)
• Second-order regression models exhibit progressive sharpening to the edge of stability [30.92413051155244]
  We show that for quadratic objectives in two dimensions, a second-order regression model exhibits progressive sharpening towards a value that differs slightly from the edge of stability. In higher dimensions, the model generically shows similar behavior, even without the specific structure of a neural network.
  arXiv  Detail & Related papers  (2022-10-10T17:21:20Z)
• Nonparametric Functional Analysis of Generalized Linear Models Under Nonlinear Constraints [0.0]
  This article introduces a novel nonparametric methodology for Generalized Linear Models. It combines the strengths of the binary regression and latent variable formulations for categorical data. It extends recently published parametric versions of the methodology and generalizes it.
  arXiv  Detail & Related papers  (2021-10-11T04:49:59Z)
• Rank-R FNN: A Tensor-Based Learning Model for High-Order Data Classification [69.26747803963907]
  Rank-R Feedforward Neural Network (FNN) is a tensor-based nonlinear learning model that imposes Canonical/Polyadic decomposition on its parameters. First, it handles inputs as multilinear arrays, bypassing the need for vectorization, and can thus fully exploit the structural information along every data dimension. We establish the universal approximation and learnability properties of Rank-R FNN, and we validate its performance on real-world hyperspectral datasets.
  arXiv  Detail & Related papers  (2021-04-11T16:37:32Z)
• LQF: Linear Quadratic Fine-Tuning [114.3840147070712]
  We present the first method for linearizing a pre-trained model that achieves comparable performance to non-linear fine-tuning. LQF consists of simple modifications to the architecture, loss function and optimization typically used for classification.
  arXiv  Detail & Related papers  (2020-12-21T06:40:20Z)
• Non-parametric Models for Non-negative Functions [48.7576911714538]
  We provide the first model for non-negative functions from the same good linear models. We prove that it admits a representer theorem and provide an efficient dual formulation for convex problems.
  arXiv  Detail & Related papers  (2020-07-08T07:17:28Z)
• Learning Stable Nonparametric Dynamical Systems with Gaussian Process Regression [9.126353101382607]
  We learn a nonparametric Lyapunov function based on Gaussian process regression from data. We prove that stabilization of the nominal model based on the nonparametric control Lyapunov function does not modify the behavior of the nominal model at training samples.
  arXiv  Detail & Related papers  (2020-06-14T11:17:17Z)

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.