An official website of the United States government
We overview several properties—old and new—of training overparameterized deep networks under the square loss. We first consider a model of the dynamics of gradient flow under the square loss in deep homogeneous rectified linear unit networks. We study the convergence to a solution with the absolute minimumρ, which is the product of the Frobenius norms of each layer weight matrix, when normalization by Lagrange multipliers is used together with weight decay under different forms of gradient descent. A main property of the minimizers that bound their expected error for a specific network architecture isρ. In particular, we derive novel norm-based bounds for convolutional layers that are orders of magnitude better than classical bounds for dense networks. Next, we prove that quasi-interpolating solutions obtained by stochastic gradient descent in the presence of weight decay have a bias toward low-rank weight matrices, which should improve generalization. The same analysis predicts the existence of an inherent stochastic gradient descent noise for deep networks. In both cases, we verify our predictions experimentally. We then predict neural collapse and its properties without any specific assumption—unlike other published proofs. Our analysis supports the idea that the advantage of deep networks relative to other classifiers is greater for problems that are appropriate for sparse deep architectures such as convolutional neural networks. The reason is that compositionally sparse target functions can be approximated well by “sparse” deep networks without incurring in the curse of dimensionality.
more »
« less
Deep Classifiers trained with the Square Loss
We overview several properties -- old and new -- of training overparametrized deep networks under the square loss. We first consider a model of the dynamics of gradient flow under the square loss in deep homogeneous ReLU networks. We study the convergence to a solution with the absolute minimum $$\rho$$, which is the product of the Frobenius norms of each layer weight matrix, when normalization by Lagrange multipliers (LM) is used together with Weight Decay (WD) under different forms of gradient descent. A main property of the minimizers that bounds their expected error {\it for a specific network architecture} is $$\rho$$. In particular, we derive novel norm-based bounds for convolutional layers that are orders of magnitude better than classical bounds for dense networks. Next we prove that quasi-interpolating solutions obtained by Stochastic Gradient Descent (SGD) in the presence of WD have a bias towards low rank weight matrices -- that, as we also explain, should improve generalization. The same analysis predicts the existence of an inherent SGD noise for deep networks. In both cases, we verify our predictions experimentally. We then predict Neural Collapse and its properties without any specific assumption -- unlike other published proofs. Our analysis supports the idea that the advantage of deep networks relative to other classifiers is greater for the problems that are appropriate for sparse deep architectures such as CNNs. The deep reason compositionally sparse target functions can be approximated well by ``sparse'' deep networks without incurring in the curse of dimensionality.
more »
« less
- Award ID(s):
- 2134108
- PAR ID:
- 10565471
- Publisher / Repository:
- Center for Brains, Minds and Machines (CBMM)
- Date Published:
- Format(s):
- Medium: X
- Institution:
- Massachusetts Institute of Technology
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We present a convergence rate analysis for biased stochastic gradient descent (SGD), where individual gradient updates are corrupted by computation errors. We develop stochastic quadratic constraints to formulate a small linear matrix inequality (LMI) whose feasible points lead to convergence bounds of biased SGD. Based on this LMI condition, we develop a sequential minimization approach to analyze the intricate trade-offs that couple stepsize selection, convergence rate, optimization accuracy, and robustness to gradient inaccuracy. We also provide feasible points for this LMI and obtain theoretical formulas that quantify the convergence properties of biased SGD under various assumptions on the loss functions.more » « less
-
We analyze deep ReLU neural networks trained with mini-batch stochastic gradient decent and weight decay. We prove that the source of the SGD noise is an implicit low rank constraint across all of the weight matrices within the network. Furthermore, we show, both theoretically and empirically, that when training a neural network using Stochastic Gradient Descent (SGD) with a small batch size, the resulting weight matrices are expected to be of small rank. Our analysis relies on a minimal set of assumptions and the neural networks may include convolutional layers, residual connections, as well as batch normalization layers.more » « less
-
Stochastic Gradient Descent (SGD) based methods have been widely used for training large-scale machine learning models that also generalize well in practice. Several explanations have been offered for this generalization performance, a prominent one being algorithmic stability [18]. However, there are no known examples of smooth loss functions for which the analysis can be shown to be tight. Furthermore, apart from the properties of the loss function, data distribution has also been shown to be an important factor in generalization performance. This raises the question: is the stability analysis of [18] tight for smooth functions, and if not, for what kind of loss functions and data distributions can the stability analysis be improved? In this paper we first settle open questions regarding tightness of bounds in the data-independent setting: we show that for general datasets, the existing analysis for convex and strongly-convex loss functions is tight, but it can be improved for non-convex loss functions. Next, we give a novel and improved data-dependent bounds: we show stability upper bounds for a large class of convex regularized loss functions, with negligible regularization parameters, and improve existing data-dependent bounds in the non-convex setting. We hope that our results will initiate further efforts to better understand the data-dependent setting under non-convex loss functions, leading to an improved understanding of the generalization abilities of deep networks.more » « less
-
Stochastic Gradient Descent (SGD) based methods have been widely used for training large-scale machine learning models that also generalize well in practice. Several explanations have been offered for this generalization performance, a prominent one being algorithmic stability Hardt et al [2016]. However, there are no known examples of smooth loss functions for which the analysis can be shown to be tight. Furthermore, apart from properties of the loss function, data distribution has also been shown to be an important factor in generalization performance. This raises the question: is the stability analysis of Hardt et al [2016] tight for smooth functions, and if not, for what kind of loss functions and data distributions can the stability analysis be improved? In this paper we first settle open questions regarding tightness of bounds in the data-independent setting: we show that for general datasets, the existing analysis for convex and strongly-convex loss functions is tight, but it can be improved for non-convex loss functions. Next, we give novel and improved data-dependent bounds: we show stability upper bounds for a large class of convex regularized loss functions, with negligible regularization parameters, and improve existing data-dependent bounds in the non-convex setting. We hope that our results will initiate further efforts to better understand the data-dependent setting under non-convex loss functions, leading to an improved understanding of the generalization abilities of deep networks.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Posted Content:
- The DOI is not currently available.
