Yoni Choukroun

I am a Computer Science Ph.D. candidate at Tel Aviv University, working in the Deep Learning Lab under the supervision of Prof. Lior Wolf. I obtained my BSc in Computer Engineering from the Technion and my M.Sc. in Computer Science from the Technion under the supervision of Prof. Ron Kimmel. I am also a team lead AI research scientist at Huawei Israel Research Center.

My research interests are centered around machine/deep learning, computer vision and numerical optimization.

Email  /  Google Scholar  /  Github  /  Linkedin

profile photo
Research
safs_small Learning Linear Block Error Correction Codes
Yoni Choukroun, Lior Wolf
ICML, 2024

We learn new binary linear block codes by cotraining the code together with a new SOTA Transformer neural decoder.

safs_small A Foundation Model for Error Correction Codes
Yoni Choukroun, Lior Wolf
ICLR, 2024

We propose a new foundational decoder that is code, length, and rate invariant and able to reach SOTA even on zero-shot scenarios.

safs_small Deep Quantum Error Correction
Yoni Choukroun, Lior Wolf
AAAI, 2024

We outperform existing solutions by overcoming several quantum ECC challenges with a deep-learning-based decoder.

safs_small Reconstructing the Hemodynamic Response Function via a Bimodal Transformer
Yoni Choukroun, Lior Golgher, Pablo Blinder, Lior Wolf
MICCAI, 2023

We learn the HRF time series via a geometry-aware transformer applied to the neurovascular modalities.

safs_small Denoising Diffusion Error Correction Codes
Yoni Choukroun, Lior Wolf
ICLR, 2023 (Notable Paper)

We modelize the channel corruption and the decoding process as a forward and reverse diffusion process.

safs_small Error Correction Code Transformer
Yoni Choukroun, Lior Wolf
Neurips, 2022

We employ an adapted Transformer in order to incorporate information about the code and to allow efficient cross-analysis of the channel's output elements.

safs_small Geometric Transformer for End-to-End Molecule Properties Prediction
Yoni Choukroun, Lior Wolf
IJCAI, 2022

We develop a novel Transformer based architecture able to learn atomic interactions and we propose a novel molecular data augmentation strategy.

safs_small Primal-Dual Sequential Subspace Optimization for Saddle-point Problems
Yoni Choukroun, Michael Zibulevski, Pavel Kisilev
NeuripsW, Optimization for Machine Learning 2020

We present a proximal subspace optimization algorithm coupled with a second-order oracle backtracking line-search for the optimization of saddle-point problems.

safs_small Low-bit quantization of neural networks for efficient inference
Yoni Choukroun, Eli Kravchik, Fan Yang, Pavel Kisilev
ICCVW, 2019

We develop post-training quantization of neural networks to up to 4 bits using an optimal MSE minimization, dual kernel refinement and differentiable finetuning of the quantizatio parameters.

safs_small Hamiltonian Operator for Spectral Shape Analysis
Yoni Choukroun, Alon Shtern, Alex Bronstein, Ron Kimmel
IEEE Transactions on Visualization and Computer Graphics, 2018

We propose the analysis and the adaptation of the Hamiltonian operator to the field of shape analysis.

safs_small Sparse approximation of 3D meshes using the spectral geometry of the Hamiltonian operator
Yoni Choukroun, Gautam Pai, Ron Kimmel
Journal of Mathematical Imaging and Vision, 2018

We propose the compression of 3D surfaces using sparse matching pursuit over the mesh dictionary built upon the Hamiltonian Operator eigenspace.

safs_small Deep learning for Decoding of Linear Codes- a Syndrome-based Approach
Amir Bennatan*, Yoni Choukroun*, Pavel Kisilev
IEEE International Symposium on Information Theory (ISIT), 2018

We present a preprocessing method of the channel output in order to make the model robust to codeword overfitting.

safs_small Schrödinger Operator for Sparse Approximation of 3D Meshes.
Yoni Choukroun, Gautam Pai, Ron Kimmel
Proceedings of the Symposium on Geometry Processing: Posters, 2017

We propose the compression of 3D meshes using sparse matching pursuit over the mesh dictionary.

safs_small Mammogram Classification and Abnormality Detection from Nonlocal Labels using Deep Multiple Instance Neural Network.
Yoni Choukroun, Ran Bakalo, Rami Ben-Ari, Ayelet Akselrod-Ballin, Ela Barkan, Pavel Kisilev
Visual Computing for Biology and Medicine, 2017

We develop a patch-based multiple instance learning method for mammogram classification.

safs_small Consistent Discretization and Minimization of the l1 Norm on Manifolds.
Alex Bronstein, Yoni Choukroun, Ron Kimmel, Matan Sela
International Conference on 3D Vision (3DV), 2016

We propose a novel IRLS approximation of the L1 on manifolds equivalent to a constrained iterative Hamiltonian eigendecomposition.

Preprints
safs_small Efficient Meta Subspace Optimization
Yoni Choukroun, Michael Katz

We propose a meta-learning based approach for the selection of the directions defining the suspace span.

safs_small Deep discriminative Latent Space for Clustering
Elad Tzoreff, Olga Kogan, Yoni Choukroun

We develop a novel unsupervised clustering method by proposing a novel discriminative encoding objective and a similarity aware clustering phase.

Patents
Computation of a saddle-point
Neural network quantization method using multiple refined quantized kernels for constrained hardware deployment
Systems and methods for automatic detection of an indication of abnormality in an anatomical image
System and a method for error correction coding using a deep neural network
Multi-thread systolic array

This page design is based on a template by Jon Barron.