Author name cluster

Alon Brutzkus

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

9 papers

2 author rows

ICML Conference 2024 Conference Paper

How Uniform Random Weights Induce Non-uniform Bias: Typical Interpolating Neural Networks Generalize with Narrow Teachers

Gon Buzaglo
Itamar Harel
Mor Shpigel Nacson
Alon Brutzkus
Nathan Srebro
Daniel Soudry

A main theoretical puzzle is why over-parameterized Neural Networks (NNs) generalize well when trained to zero loss (i. e. , so they interpolate the data). Usually, the NN is trained with Stochastic Gradient Descent (SGD) or one of its variants. However, recent empirical work examined the generalization of a random NN that interpolates the data: the NN was sampled from a seemingly uniform prior over the parameters, conditioned on that the NN perfectly classifying the training set. Interestingly, such a NN sample typically generalized as well as SGD-trained NNs. We prove that such a random NN interpolator typically generalizes well if there exists an underlying narrow “teacher NN" that agrees with the labels. Specifically, we show that such a ‘flat’ prior over the NN parametrization induces a rich prior over the NN functions, due to the redundancy in the NN structure. In particular, this creates a bias towards simpler functions, which require less relevant parameters to represent — enabling learning with a sample complexity approximately proportional to the complexity of the teacher (roughly, the number of non-redundant parameters), rather than the student’s.

Details

ICML Conference 2022 Conference Paper

Efficient Learning of CNNs using Patch Based Features

Alon Brutzkus
Amir Globerson
Eran Malach
Alon Regev Netser
Shai Shalev-Shwartz

Recent work has demonstrated the effectiveness of using patch based representations when learning from image data. Here we provide theoretical support for this observation, by showing that a simple semi-supervised algorithm that uses patch statistics can efficiently learn labels produced by a one-hidden-layer Convolutional Neural Network (CNN). Since CNNs are known to be computationally hard to learn in the worst case, our analysis holds under some distributional assumptions. We show that these assumptions are necessary and sufficient for our results to hold. We verify that the distributional assumptions hold on real-world data by experimenting on the CIFAR-10 dataset, and find that the analyzed algorithm outperforms a vanilla one-hidden-layer CNN. Finally, we demonstrate that by running the algorithm in a layer-by-layer fashion we can build a deep model which gives further improvements, hinting that this method provides insights about the behavior of deep CNNs.

Details

UAI Conference 2022 Conference Paper

On the inductive bias of neural networks for learning read-once DNFs

Ido Bronstein
Alon Brutzkus
Amir Globerson

Learning functions over Boolean variables is a fundamental problem in machine learning. But not much is known about learning such functions using neural networks. Here we focus on learning read-once disjunctive normal forms (DNFs) under the uniform distribution with a convex neural network and gradient methods. We first observe empirically that gradient methods converge to compact solutions with neurons that are aligned with the terms of the DNF. This is despite the fact that there are many zero training error networks that do not have this property. Thus, the learning process has a clear inductive bias towards such logical formulas. Following recent results which connect the inductive bias of gradient flow (GF) to Karush-Kuhn-Tucker (KKT) points of minimum norm problems, we study these KKT points in our setting. We prove that zero training error solutions that memorize training points are not KKT points and therefore GF cannot converge to them. On the other hand, we prove that globally optimal KKT points correspond exactly to networks that are aligned with the DNF terms. These results suggest a strong connection between the inductive bias of GF and solutions that align with the DNF. We conclude with extensive experiments which verify our findings.

Details

NeurIPS Conference 2021 Conference Paper

A Theoretical Analysis of Fine-tuning with Linear Teachers

Gal Shachaf
Alon Brutzkus
Amir Globerson

Fine-tuning is a common practice in deep learning, achieving excellent generalization results on downstream tasks using relatively little training data. Although widely used in practice, it is not well understood theoretically. Here we analyze the sample complexity of this scheme for regression with linear teachers in several settings. Intuitively, the success of fine-tuning depends on the similarity between the source tasks and the target task. But what is the right way of measuring this similarity? We show that the relevant measure has to do with the relation between the source task, the target task and the covariance structure of the target data. In the setting of linear regression, we show that under realistic settings there can be substantial sample complexity reduction when the above measure is low. For deep linear regression, we propose a novel result regarding the inductive bias of gradient-based training when the network is initialized with pretrained weights. Using this result we show that the similarity measure for this setting is also affected by the depth of the network. We conclude with results on shallow ReLU models, and analyze the dependence of sample complexity there on source and target tasks. We empirically demonstrate our results for both synthetic and realistic data.

PDF Details

UAI Conference 2021 Conference Paper

An optimization and generalization analysis for max-pooling networks

Alon Brutzkus
Amir Globerson

Max-Pooling operations are a core component of deep learning architectures. In particular, they are part of most convolutional architectures used in machine vision, since pooling is a natural approach to pattern detection problems. However, these architectures are not well understood from a theoretical perspective. For example, we do not understand when they can be globally optimized, and what is the effect of over-parameterization on generalization. Here we perform a theoretical analysis of a convolutional max-pooling architecture, proving that it can be globally optimized, and can generalize well even for highly over-parameterized models. Our analysis focuses on a data generating distribution inspired by pattern detection problem, where a “discriminative” pattern needs to be detected among “spurious” patterns. We empirically validate that CNNs significantly outperform fully connected networks in our setting, as predicted by our theoretical results.

Details

ICML Conference 2021 Conference Paper

Towards Understanding Learning in Neural Networks with Linear Teachers

Roei Sarussi
Alon Brutzkus
Amir Globerson

Can a neural network minimizing cross-entropy learn linearly separable data? Despite progress in the theory of deep learning, this question remains unsolved. Here we prove that SGD globally optimizes this learning problem for a two-layer network with Leaky ReLU activations. The learned network can in principle be very complex. However, empirical evidence suggests that it often turns out to be approximately linear. We provide theoretical support for this phenomenon by proving that if network weights converge to two weight clusters, this will imply an approximately linear decision boundary. Finally, we show a condition on the optimization that leads to weight clustering. We provide empirical results that validate our theoretical analysis.

Details

ICML Conference 2019 Conference Paper

Low Latency Privacy Preserving Inference

Alon Brutzkus
Ran Bachrach
Oren Elisha

When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than 10\times improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of \sim0. 16 seconds. We demonstrate the efficacy of our methods on several computer vision tasks.

Details

ICML Conference 2019 Conference Paper

Why do Larger Models Generalize Better? A Theoretical Perspective via the XOR Problem

Alon Brutzkus
Amir Globerson

Empirical evidence suggests that neural networks with ReLU activations generalize better with over-parameterization. However, there is currently no theoretical analysis that explains this observation. In this work, we provide theoretical and empirical evidence that, in certain cases, overparameterized convolutional networks generalize better than small networks because of an interplay between weight clustering and feature exploration at initialization. We demonstrate this theoretically for a 3-layer convolutional neural network with max-pooling, in a novel setting which extends the XOR problem. We show that this interplay implies that with overparamterization, gradient descent converges to global minima with better generalization performance compared to global minima of small networks. Empirically, we demonstrate these phenomena for a 3-layer convolutional neural network in the MNIST task.

Details

ICML Conference 2017 Conference Paper

Globally Optimal Gradient Descent for a ConvNet with Gaussian Inputs

Alon Brutzkus
Amir Globerson

Deep learning models are often successfully trained using gradient descent, despite the worst case hardness of the underlying non-convex optimization problem. The key question is then under what conditions can one prove that optimization will succeed. Here we provide a strong result of this kind. We consider a neural net with one hidden layer and a convolutional structure with no overlap and a ReLU activation function. For this architecture we show that learning is NP-complete in the general case, but that when the input distribution is Gaussian, gradient descent converges to the global optimum in polynomial time. To the best of our knowledge, this is the first global optimality guarantee of gradient descent on a convolutional neural network with ReLU activations.

Details