Search Results for author: David Burshtein

Found 8 papers, 2 papers with code

Semi-Supervised Variational Inference over Nonlinear Channels

no code implementations21 Sep 2023 David Burshtein, Eli Bery

Deep learning methods for communications over unknown nonlinear channels have attracted considerable interest recently.

Meta-Learning Variational Inference

Unsupervised Linear and Nonlinear Channel Equalization and Decoding using Variational Autoencoders

no code implementations21 May 2019 Avi Caciularu, David Burshtein

We first consider the reconstruction of uncoded data symbols transmitted over a noisy linear intersymbol interference (ISI) channel, with an unknown impulse response, without using pilot symbols.

Variational Inference

Blind Channel Equalization using Variational Autoencoders

no code implementations5 Mar 2018 Avi Caciularu, David Burshtein

A new maximum likelihood estimation approach for blind channel equalization, using variational autoencoders (VAEs), is introduced.

Near Maximum Likelihood Decoding with Deep Learning

no code implementations8 Jan 2018 Eliya Nachmani, Yaron Bachar, Elad Marciano, David Burshtein, Yair Be'ery

The proposed decoder is based on the neural Belief Propagation algorithm and the Automorphism Group.

Deep Learning Methods for Improved Decoding of Linear Codes

2 code implementations21 Jun 2017 Eliya Nachmani, Elad Marciano, Loren Lugosch, Warren J. Gross, David Burshtein, Yair Beery

Furthermore, we demonstrate that the neural belief propagation decoder can be used to improve the performance, or alternatively reduce the computational complexity, of a close to optimal decoder of short BCH codes.

Simplified End-to-End MMI Training and Voting for ASR

no code implementations30 Mar 2017 Lior Fritz, David Burshtein

A simplified speech recognition system that uses the maximum mutual information (MMI) criterion is considered.

General Classification Language Modelling +2

RNN Decoding of Linear Block Codes

no code implementations24 Feb 2017 Eliya Nachmani, Elad Marciano, David Burshtein, Yair Be'ery

We also demonstrate improved performance over belief propagation on sparser Tanner graph representations of the codes.

Learning to Decode Linear Codes Using Deep Learning

2 code implementations16 Jul 2016 Eliya Nachmani, Yair Beery, David Burshtein

A novel deep learning method for improving the belief propagation algorithm is proposed.

Cannot find the paper you are looking for? You can Submit a new open access paper.