Amit Daniely

is this you? claim profile

1 follower
Google ∙ Hebrew University of Jerusalem

From Local Pseudorandom Generators to Hardness of Learning

We prove hardness-of-learning results under a well-studied assumption on...
01/20/2021 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Most ReLU Networks Suffer from ℓ^2 Adversarial Perturbations

We consider ReLU networks with random weights, in which the dimension de...
10/28/2020 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Hardness of Learning Neural Networks with Natural Weights

Neural networks are nowadays highly successful despite strong hardness r...
06/05/2020 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Memorizing Gaussians with no over-parameterizaion via gradient decent on neural networks

We prove that a single step of gradient decent over depth two network, w...
03/28/2020 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Learning Parities with Neural Networks

In recent years we see a rapidly growing line of research which shows le...
02/18/2020 ∙ by Amit Daniely, et al. ∙ 2 ∙ share

read it
On the Complexity of Minimizing Convex Finite Sums Without Using the Indices of the Individual Functions

Recent advances in randomized incremental methods for minimizing L-smoot...
02/09/2020 ∙ by Yossi Arjevani, et al. ∙ 5 ∙ share

read it
Neural Networks Learning and Memorization with (almost) no Over-Parameterization

Many results in recent years established polynomial time learnability of...
11/22/2019 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Generalization Bounds for Neural Networks via Approximate Description Length

We investigate the sample complexity of networks with bounds on the magn...
10/13/2019 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
The Implicit Bias of Depth: How Incremental Learning Drives Generalization

A leading hypothesis for the surprising generalization of neural network...
09/26/2019 ∙ by Daniel Gissin, et al. ∙ 0 ∙ share

read it
On the Optimality of Trees Generated by ID3

Since its inception in the 1980s, ID3 has become one of the most success...
07/11/2019 ∙ by Alon Brutzkus, et al. ∙ 0 ∙ share

read it
ID3 Learns Juntas for Smoothed Product Distributions

In recent years, there are many attempts to understand popular heuristic...
06/20/2019 ∙ by Alon Brutzkus, et al. ∙ 0 ∙ share

read it
Competitive ratio versus regret minimization: achieving the best of both worlds

We consider online algorithms under both the competitive ratio criteria ...
04/07/2019 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Learning without Interaction Requires Separation

One of the key resources in large-scale learning systems is the number o...
09/24/2018 ∙ by Amit Daniely, et al. ∙ 2 ∙ share

read it
Planning and Learning with Stochastic Action Sets

In many practical uses of reinforcement learning (RL) the set of actions...
05/07/2018 ∙ by Craig Boutilier, et al. ∙ 0 ∙ share

read it
Learning with Rules

Complex classifiers may exhibit "embarassing" failures in cases that wou...
03/08/2018 ∙ by Deborah Cohen, et al. ∙ 0 ∙ share

read it
SGD Learns the Conjugate Kernel Class of the Network

We show that the standard stochastic gradient decent (SGD) algorithm is ...
02/27/2017 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Depth Separation for Neural Networks

Let f:S^d-1×S^d-1→S be a function of the form f(x,x') = g(〈x,x'〉) for g:...
02/27/2017 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it
Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity

We develop a general duality between neural networks and compositional k...
02/18/2016 ∙ by Amit Daniely, et al. ∙ 0 ∙ share

read it

Amit Daniely

Featured Co-authors

Yishay Mansour

Stefanie Jegelka

Dale Schuurmans

Craig Boutilier

Amir Globerson

Shai Shalev-Shwartz

Vitaly Feldman

Avinatan Hassidim

Yoram Singer

Gal Elidan

Eran Malach

From Local Pseudorandom Generators to Hardness of Learning

Most ReLU Networks Suffer from ℓ^2 Adversarial Perturbations

Hardness of Learning Neural Networks with Natural Weights

Memorizing Gaussians with no over-parameterizaion via gradient decent on neural networks

Learning Parities with Neural Networks

On the Complexity of Minimizing Convex Finite Sums Without Using the Indices of the Individual Functions

Neural Networks Learning and Memorization with (almost) no Over-Parameterization

Generalization Bounds for Neural Networks via Approximate Description Length

The Implicit Bias of Depth: How Incremental Learning Drives Generalization

On the Optimality of Trees Generated by ID3

ID3 Learns Juntas for Smoothed Product Distributions

Competitive ratio versus regret minimization: achieving the best of both worlds

Learning without Interaction Requires Separation

Planning and Learning with Stochastic Action Sets

Learning with Rules

SGD Learns the Conjugate Kernel Class of the Network

Depth Separation for Neural Networks

Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity