1 code implementation • 12 Feb 2024 • Noam Razin, Yotam Alexander, Edo Cohen-Karlik, Raja Giryes, Amir Globerson, Nadav Cohen
This paper theoretically studies the implicit bias of policy gradient in terms of extrapolation to unseen initial states.
no code implementations • 28 Jan 2024 • Nadav Cohen, Itzik Klein
Autonomous underwater vehicles are specialized platforms engineered for deep underwater operations.
1 code implementation • 18 Jan 2024 • Nadav Cohen, Itzik Klein
In this paper, we derive and introduce A-KIT, an adaptive Kalman-informed transformer to learn the varying process noise covariance online.
no code implementations • 22 Jun 2023 • Nadav Cohen, Itzik Klein
The latter is done by learning some of the fusion filter parameters.
1 code implementation • 20 Mar 2023 • Yotam Alexander, Nimrod De La Vega, Noam Razin, Nadav Cohen
Focusing on locally connected neural networks (a prevalent family of architectures that includes convolutional and recurrent neural networks as well as local self-attention models), we address this problem by adopting theoretical tools from quantum physics.
no code implementations • 22 Dec 2022 • Nadav Cohen, Zeev Yampolsky, Itzik Klein
Our ST-BeamsNet estimated the AUV velocity vector with an 8. 547% speed error, which is 26% better than the MA approach.
1 code implementation • NeurIPS 2023 • Noam Razin, Tom Verbin, Nadav Cohen
Formalizing strength of interactions through an established measure known as separation rank, we quantify the ability of certain GNNs to model interaction between a given subset of vertices and its complement, i. e. between the sides of a given partition of input vertices.
no code implementations • 25 Oct 2022 • Edo Cohen-Karlik, Itamar Menuhin-Gruman, Raja Giryes, Nadav Cohen, Amir Globerson
Overparameterization in deep learning typically refers to settings where a trained neural network (NN) has representational capacity to fit the training data in many ways, some of which generalize well, while others do not.
no code implementations • 22 Oct 2022 • Nadav Cohen, Govind Menon, Zsolt Veraszto
The deep linear network (DLN) is a model for implicit regularization in gradient based optimization of overparametrized learning architectures.
no code implementations • 20 Oct 2022 • Nadav Cohen, Itzik Klein
In such conditions, the vehicle's velocity vector could not be estimated leading to a navigation solution drift and in some situations the AUV is required to abort the mission and return to the surface.
no code implementations • 27 Jun 2022 • Nadav Cohen, Itzik Klein
Both simulation and sea experiments were made to validate the proposed learning approach relative to the model-based approach.
1 code implementation • 7 Mar 2022 • Nadav Cohen, Yael Newman, Ariel Shamir
In this paper, we tackle the problem of semantic segmentation of artistic paintings, an even more challenging task because of a much larger diversity in colors, textures, and shapes and because there are no ground truth annotations available for segmentation.
no code implementations • 9 Feb 2022 • Edo Cohen-Karlik, Avichai Ben David, Nadav Cohen, Amir Globerson
When using recurrent neural networks (RNNs) it is common practice to apply trained models to sequences longer than those seen in training.
1 code implementation • 27 Jan 2022 • Noam Razin, Asaf Maman, Nadav Cohen
In the pursuit of explaining implicit regularization in deep learning, prominent focus was given to matrix and tensor factorizations, which correspond to simplified neural networks.
1 code implementation • NeurIPS 2021 • Omer Elkabetz, Nadav Cohen
The extent to which it represents gradient descent is an open question in the theory of deep learning.
1 code implementation • 19 Feb 2021 • Noam Razin, Asaf Maman, Nadav Cohen
Recent efforts to unravel the mystery of implicit regularization in deep learning have led to a theoretical focus on matrix factorization -- matrix completion via linear neural network.
1 code implementation • NeurIPS 2020 • Noam Razin, Nadav Cohen
Mathematically characterizing the implicit regularization induced by gradient-based optimization is a longstanding pursuit in the theory of deep learning.
1 code implementation • NeurIPS 2019 • Sanjeev Arora, Nadav Cohen, Wei Hu, Yuping Luo
Efforts to understand the generalization mystery in deep learning have led to the belief that gradient-based optimization induces a form of implicit regularization, a bias towards models of low "complexity."
no code implementations • ICLR 2019 • Sanjeev Arora, Nadav Cohen, Noah Golowich, Wei Hu
We analyze speed of convergence to global optimum for gradient descent training a deep linear neural network (parameterized as $x \mapsto W_N W_{N-1} \cdots W_1 x$) by minimizing the $\ell_2$ loss over whitened data.
no code implementations • CVPR 2018 • Assaf Shocher, Nadav Cohen, Michal Irani
On such images, our method outperforms SotA CNN-based SR methods, as well as previous unsupervised SR methods.
no code implementations • 26 Mar 2018 • Yoav Levine, Or Sharir, Nadav Cohen, Amnon Shashua
Modern deep learning has enabled unprecedented achievements in various domains.
1 code implementation • ICML 2018 • Sanjeev Arora, Nadav Cohen, Elad Hazan
The effect of depth on optimization is decoupled from expressiveness by focusing on settings where additional layers amount to overparameterization - linear neural networks, a well-studied model.
7 code implementations • 17 Dec 2017 • Assaf Shocher, Nadav Cohen, Michal Irani
On such images, our method outperforms SotA CNN-based SR methods, as well as previous unsupervised SR methods.
Ranked #46 on Image Super-Resolution on BSD100 - 4x upscaling
no code implementations • 5 May 2017 • Nadav Cohen, Or Sharir, Yoav Levine, Ronen Tamari, David Yakira, Amnon Shashua
Expressive efficiency refers to the ability of a network architecture to realize functions that require an alternative architecture to be much larger.
no code implementations • ICLR 2018 • Yoav Levine, David Yakira, Nadav Cohen, Amnon Shashua
This description enables us to carry a graph-theoretic analysis of a convolutional network, with which we demonstrate a direct control over the inductive bias of the deep network via its channel numbers, that are related to the min-cut in the underlying graph.
no code implementations • ICLR 2018 • Nadav Cohen, Ronen Tamari, Amnon Shashua
By introducing and analyzing the concept of mixed tensor decompositions, we prove that interconnecting dilated convolutional networks can lead to expressive efficiency.
2 code implementations • 13 Oct 2016 • Or Sharir, Ronen Tamari, Nadav Cohen, Amnon Shashua
Other methods, based on arithmetic circuits and sum-product networks, do allow tractable marginalization, but their performance is challenged by the need to learn the structure of a circuit.
1 code implementation • 22 May 2016 • Nadav Cohen, Amnon Shashua
In addition to analyzing deep networks, we show that shallow ones support only linear separation ranks, and by this gain insight into the benefit of functions brought forth by depth - they are able to efficiently model strong correlation under favored partitions of the input.
no code implementations • 1 Mar 2016 • Nadav Cohen, Amnon Shashua
Second, and more importantly, we show that depth efficiency is weaker with convolutional rectifier networks than it is with convolutional arithmetic circuits.
no code implementations • 16 Sep 2015 • Nadav Cohen, Or Sharir, Amnon Shashua
In this work we derive a deep network architecture based on arithmetic circuits that inherently employs locality, sharing and pooling.
no code implementations • CVPR 2016 • Nadav Cohen, Or Sharir, Amnon Shashua
We present a deep layered architecture that generalizes convolutional neural networks (ConvNets).
1 code implementation • 3 Oct 2014 • Nadav Cohen, Amnon Shashua
We present a deep layered architecture that generalizes classical convolutional neural networks (ConvNets).