Introduction

Welcome to the homepage of the NVIDIA Research Israel AI Lab. Our research spans algorithms, theory and applications of deep learning, with a focus on computer vision and reinforcement learning. We are particularly interested in perception, action, and reasoning (PAR), and their intersection. The lab is led by Professor Gal Chechik. See recent projects and current team members.

News

May 2023

The paper “CALM: Conditional Adversarial Latent Models for Directable Virtual Characters” by Chen Tessler, Yoni Kasten, Yunrong Guo, Shie Mannor, Gal Chechik and Xue Bin Peng was accepted to SIGGRAPH 2023, Project page
The paper “Key-Locked Rank One Editing for Text-to-Image Personalization” by Yoad Tewel, Rinon Gal, Gal Chechik and Yuval Atzmon was accepted to SIGGRAPH 2023, Project page
The paper “Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Models” by Rinon Gal, Moab Arar, Yuval Atzmon, Amit H. Bermano, Gal Chechik and Daniel Cohen-Or was accepted to SIGGRAPH 2023, Project page
The paper “Learning to Initiate and Reason in Event-Driven Cascading Processes” by Yuval Atzmon⁕, Eli Meirom⁕, Shie Mannor and Gal Chechik was accepted to ICML 2023, Project page

February 2023

The paper “AutoScratch: ML-Optimized GPU Cache Management” by Yaosheng Fu, Evgeny Bolotin, Aamer Jaleel, Gal Dalal, Shie Mannor, Jacob Subag, Noam Korem, Michael Behar and David Nellans was accepted to MLSys 2023, (Paper link will be published soon)

January 2023

The paper “Never Worse, Mostly Better: Stable Policy Improvement in Deep Reinforcement Learning” by Pranav Khanna, Guy Tennenholtz, Nadav Merlis, Shie Mannor and Chen Tessler was accepted to AAMAS 2023, Arxiv

November 2022

The paper “Planning and Learning with Adaptive Lookahead” by Aviv Rosenberg, Assaf Hallak, Shie Mannor, Gal Chechik, and Gal Dalal was accepted to AAAI 2023, PDF

June 2022

A new paper “Reinforcement Learning with a Terminator” by Guy Tennenholtz, Nadav Merlis, Lior Shani, Shie Mannor, Uri Shalit, Gal Chechik, Assaf Hallak, Gal Dalal. Arxiv

May 2022

The paper “Improve Agents without Retraining: Parallel Tree Search with Off-Policy Correction” by Assaf Hallak, Gal Dalal, Steven Dalton, Iuri Frosio, Shie Mannor, and Gal Chechik was accepted to RLDM 2022, PDF
The paper “Planning and Learning with Adaptive Lookahead” by Aviv Rosenberg, Assaf Hallak, Shie Mannor, Gal Chechik, and Gal Dalal was accepted to RLDM 2022, PDF
The paper “Optimizing Tensor Network Contraction Using Reinforcement Learning” by Eli Meirom, Haggai Maron, Shie Mannor, and Gal Chechik was accepted to ICML 2022, PDF
The paper “Multi-Task Learning as a Bargaining Game” by Aviv Navon, Aviv Shamsian, Idan Achituve, Haggai Maron, Kenji Kawaguchi, Gal Chechik, Ethan Fetaya was accepted to ICML 2022, PDF
A new paper “A Simple and Universal Rotation Equivariant Point-cloud Network” by Ben Finkelshtein, Chaim Baskin, Haggai Maron, Nadav Dym. PDF
A new paper “Sign and Basis Invariant Networks for Spectral Graph Representation Learning” by Derek Lim, Joshua Robinson, Lingxiao Zhao, Tess Smidt, Suvrit Sra, Haggai Maron, Stefanie Jegelka. PDF
A new paper “Understanding and Extending Subgraph GNNs by Rethinking Their Symmetries” by Fabrizio Frasca, Beatrice Bevilacqua, Michael M. Bronstein, Haggai Maron. PDF

April 2022

The workshop “3D Perception for Autonomous Driving” is accepted to ECCV 2022. Link
A new paper “Text2LIVE: Text-Driven Layered Image and Video Editing” by Omer Bar-Tal, Dolev Ofri-Amar, Rafail Fridman, Yoni Kasten, and Tali Dekel. Arxiv, Project Page

March 2022

“Optimizing Tensor Network Contraction Using Reinforcement Learning” by Eli Meirom, Haggai Maron, Shie Mannor, and Gal Chechik was accepted as an oral to RLDM 2022, PDF
“Learning to reason about and to act on physical cascading events” by Yuval Atzmon, Eli A. Meirom, Shie Mannor, and Gal Chechik was accepted to both RLDM 2022 and ICLR 2022 OSC workshop Arxiv.
“Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks” by Eli A. Meirom, Haggai Maron, Shie Mannor, Gal Chechik, was accepted to RLDM 2022, PDF.
A new paper " “This is my unicorn, Fluffy”: Personalizing frozen vision-language representations " by Niv Cohen, Rinon Gal, Eli A. Meirom, Gal Chechik, Yuval Atzmon. Arxiv

Feb 2022

The paper “Equivariant Subgraph Aggregation Networks” by Beatrice Bevilacqua, Fabrizio Frasca, Derek Lim, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron, was accepted as a spotlight presentation to ICLR 2022 , PDF.
A new paper “Optimizing Tensor Network Contraction Using Reinforcement Learning” by Eli Meirom, Haggai Maron, Shie Mannor, Gal Chechik. Arxiv
A new paper “Multi-Task Learning as a Bargaining Game” by Aviv Navon, Aviv Shamsian, Idan Achituve, Haggai Maron, Kenji Kawaguchi, Gal Chechik, Ethan Fetaya. Arxiv
A new paper “Federated Learning with Heterogeneous Architectures using Graph HyperNetworks” by Or Litany, Haggai Maron, David Acuna, Jan Kautz, Gal Chechik, Sanja Fidler. Arxiv
A new paper “Learning to reason about and to act on physical cascading events” by Yuval Atzmon, Eli A. Meirom, Shie Mannor, Gal Chechik. Arxiv

Dec 2021

Shie Mannor was elected as an IEEE Fellow.

Nov 2021

A new paper “StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generators” by Rinon Gal, Or Patashnik, Haggai Maron, Gal Chechik, Daniel Cohen-Or. Arxiv Code , Project Page
Gal presented our work on perception reasoning and action at GTC 2021, Video.
Shie presented our work on accelerated tree search with off-plicycorrection at GTC 2021, Video.
The paper “Reinforcement Learning for Datacenter Congestion Control” by Chen Tessler, Yuval Shpigelman, Gal Dalal, Amit Mendelbaum, Shie Mannor, Gal Chechik was accepted to IAAI 2022. Arxiv
Gal presented work on federated learning VIDEO

Oct 2021

The paper “Improve Agents without Retraining: Parallel Tree Search with Off-Policy Correction” by Assaf Hallak, Gal Dalal, Steven Dalton, Iuri Frosio, Shie Mannor, Gal Chechik, was accepted to NeurIPS 2021, PDF.
The paper “Sim and Real: Better Together” by Shirli Di-Castro, Dotan Di Castro, Shie Mannor, was accepted to NeurIPS 2021, PDF.
The paper “Personalized Federated Learning With Gaussian Processes” by Idan Achituve, Aviv Shamsian, Aviv Navon, Gal Chechik, Ethan Fetaya, was accepted to NeurIPS 2021 PDF
We coorganize the Israeli Geometric deep learning day 2021, Schedule

Sep 2021

New paper submitted to ICLR “On Covariate Shift of Latent Confounders in Imitation and Reinforcement Learning” by Guy Tennenholtz, Assaf Hallak, Gal Dalal, Shie Mannor, Gal Chechik, Uri Shalit, PDF.
New paper submitted to ICLR “Equivariant Subgraph Aggregation Networks” by Beatrice Bevilacqua, Fabrizio Frasca, Derek Lim, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron , PDF.

Jul 2021

A new paper “Improve Agents without Retraining: Parallel Tree Search with Off-Policy Correction” by Assaf Hallak, Gal Dalal, Steven Dalton, Iuri Frosio, Shie Mannor, Gal Chechik, was accepted to NeurIPS 2021, PDF.

May 2021

The paper “Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks” by Eli A. Meirom, Haggai Maron, Shie Mannor, Gal Chechik, was accepted to ICML 2021, PDF.
The paper “From Local Structures to Size Generalization in Graph Neural Networks” by Gilad Yehudai, Ethan Fetaya, Eli Meirom, Gal Chechik, Haggai Maron, was accepted to ICML 2021, PDF.
The paper “Personalized Federated Learning using Hypernetworks” by Aviv Shamsian, Aviv Navon, Ethan Fetaya, Gal Chechik, was accepted to ICML 2021, PDF.
The paper “GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning” by Idan Achituve, Aviv Navon, Yochai Yemini, Gal Chechik, Ethan Fetaya, was accepted to ICML 2021, PDF.
The paper “Compositional Video Synthesis with Action Graphs” by Amir Bar, Roei Herzig, Xiaolong Wang, Gal Chechik, Trevor Darrell, Amir Globerson, was accepted to ICML 2021, PDF Slides.
The paper “Detecting Rewards Deterioration in Episodic Reinforcement Learning” by “Ido Greenberg, Shie Mannor” was accpted to ICML 2021 PDF
The paper “Known unknowns: Learning novel concepts using reasoning-by-elimination” by Harsh Agrawal, Eli A. Meirom, Yuval Atzmon, Shie Mannor, and Gal Chechik was accepted as an oral to UAI 2021 PDF
New paper by Assaf Hallak, Gal Dalal, Steven Dalton, Iuri Frosio, Shie Mannor, Gal Chechik. “Improve Agents without Retraining: Parallel Tree Search with Off-Policy Correction”, PDF.

Apr 2021

We coorganize the Israeli Reinforcement Learning day, Schedule

Feb 2021

New paper by Chen Tessler, Yuval Shpigelman, Gal Dalal, Amit Mandelbaum, Doron Haritan Kazakov, Benjamin Fuhrer, Gal Chechik, Shie Mannor. Reinforcement Learning for Datacenter Congestion Control, PDF.
New paper by Sangho Lee, Jiwan Chung, Youngjae Yu, Gunhee Kim, Thomas Breuel, Gal Chechik, Yale Song. Automatic Curation of Large-Scale Datasets for Audio-Visual Representation Learning PDF

Jan 2021

The paper “Learning the Pareto front with hypernetworks” by Aviv Navon, Aviv Shamsian, Gal Chechik, Ethan Fetaya, was accepted to ICLR 2021, PDF.
The paper “Auxiliary Learning by Implicit Differentiation” by Aviv Navon, Idan Achituve, Haggai Maron, Gal Chechik, Ethan Fetaya, was accepted to ICLR 2021, PDF.
The paper “On The Universality of Rotation Equivariant Point cloud Networks” by Nadav Dym, Haggai Maron was accepted to ICLR 2021, PDF.
The paper “Acting in Delayed Environments with Non-Stationary Markov Policies” by Esther Derman, Gal Dalal, Shie Mannor, was accepted to ICLR 2021, PDF.
The paper “Optimizing Memory Placement using Evolutionary Graph Reinforcement Learning” by Shauharda Khadka, Estelle Aflalo, Mattias Mardar, Avrech Ben-David, Santiago Miret, Shie Mannor, Tamir Hazan, Hanlin Tang, Somdeb Majumdar, was accepted to ICLR 2021.
The paper “Reinforcement Learning with Trajectory Feedback” by Yonathan Efroni, Nadav Merlis, Shie Mannor, was accepted to AAAI 2021.
The paper “Lenient Regret for Multi-Armed Bandits” by N. Merlis and S. Mannor, was accepted to AAAI 2021.

Dec 2020

The paper “A causal view of compositional zero-shot recognition” by Y. Atzmon,F. Kreuk, Uri Shalit, G. Chechik, was presented as a NeurIPS spotlight, PDF.

Nov 2020

New paper by Gilad Yehudai, Ethan Fetaya, Eli Meirom, Gal Chechik, Haggai Maron. On Size Generalization in Graph Neural Networks, PDF.
New paper by Aviv Navon, Idan Achituve, Haggai Maron, Gal Chechik, Ethan Fetaya, Auxiliary Learning by Implicit Differentiation, PDF.
New paper by Nadav Dym, Haggai Maron, On The Universality of Rotation Equivariant Point cloud Networks, PDF.
new paper Gal Dalal, Esther Derman, Shie Mannor. Acting in Delayed Environments with Non-Stationary Markov Policies.

Oct 2020

The paper “From Generalized zero-shot learning to long-tail with class descriptors”, by D. Samuel,Y. Atzmon, G. Chechik, PDF, was accepted to WACV 2021.
The paper “Self-supervised learning for domain adaptation on point clouds”, by I. Achituv, H. Maron, G. Chechik, PDF, was accepted to WACV 2021
NVIDIA teaches machines to communicate: YNET (Hebrew)

July 2020