no code implementations • 9 Feb 2024 • Gon Buzaglo, Itamar Harel, Mor Shpigel Nacson, Alon Brutzkus, Nathan Srebro, Daniel Soudry
We prove that such a random NN interpolator typically generalizes well if there exists an underlying narrow ``teacher NN" that agrees with the labels.
no code implementations • 29 Sep 2021 • Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz
Convolutional networks (CNN) are computationally hard to learn.
no code implementations • NeurIPS 2021 • Gal Shachaf, Alon Brutzkus, Amir Globerson
We further present results on shallow ReLU models, and analyze the dependence of sample complexity there on source and target tasks.
no code implementations • 7 Jan 2021 • Roei Sarussi, Alon Brutzkus, Amir Globerson
Can a neural network minimizing cross-entropy learn linearly separable data?
no code implementations • 1 Jan 2021 • Ido Bronstein, Alon Brutzkus, Amir Globerson
Here we focus on this setting where the functions are learned by a convex neural network and gradient descent.
no code implementations • 1 Jan 2021 • Alon Brutzkus, Amir Globerson
In this work, we consider a simplified image classification task where images contain orthogonal patches and are learned with a 3-layer overparameterized convolutional network and stochastic gradient descent (SGD).
no code implementations • 22 Feb 2020 • Alon Brutzkus, Amir Globerson
Max-Pooling operations are a core component of deep learning architectures.
no code implementations • 11 Jul 2019 • Alon Brutzkus, Amit Daniely, Eran Malach
Since its inception in the 1980s, ID3 has become one of the most successful and widely used algorithms for learning decision trees.
no code implementations • 20 Jun 2019 • Alon Brutzkus, Amit Daniely, Eran Malach
In recent years, there are many attempts to understand popular heuristics.
no code implementations • ICLR 2019 • Alon Brutzkus, Amir Globerson
Empirical evidence suggests that neural networks with ReLU activations generalize better with over-parameterization.
1 code implementation • ICLR 2019 • Alon Brutzkus, Oren Elisha, Ran Gilad-Bachrach
When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity.
no code implementations • ICLR 2019 • Alon Brutzkus, Amir Globerson
Empirical evidence suggests that neural networks with ReLU activations generalize better with over-parameterization.
no code implementations • ICLR 2018 • Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz
Neural networks exhibit good generalization behavior in the over-parameterized regime, where the number of network parameters exceeds the number of observations.
no code implementations • ICML 2017 • 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.