Deep Learning for Holistic Inference

Informazioni relative al progetto

HOLI

ID dell’accordo di sovvenzione: 819080

DOI

10.3030/819080

Progetto chiuso

Data della firma CE 17 Gennaio 2019

Data di avvio 1 Febbraio 2019

Data di completamento 31 Gennaio 2025

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale

€ 1 932 500,00

Contributo UE

€ 1 932 500,00

Coordinato da

TEL AVIV UNIVERSITY
Israel

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Pubblicazioni

Conference proceedings (30)

Finding Visual Task Vectors

Autori: Alberto Hojel, Yutong Bai, Trevor Darrell, Amir Globerson, Amir Bar
Pubblicato in: ECCV24, 2024
Editore: Springer

Provable Benefits of Complex Parameterizations for Structured State Space Models

Autori: Yuval Ran-Milo, Eden Lumbroso, Edo Cohen-Karlik, Raja Giryes, Amir Globerson, Nadav Cohen
Pubblicato in: 2024
Editore: NeurIPS

Compositional Video Synthesis with Action Graphs

Autori: Bar, Amir; Herzig, Roei; Wang, Xiaolong; Chechik, Gal; Darrell, Trevor; Globerson, Amir
Pubblicato in: International Conference on Machine Learning (ICML, 2021), Numero 1, 2021
Editore: The International Machine Learning Society

What Are You Token About? Dense Retrieval as Distributions Over the Vocabulary

Autori: Ori Ram, Liat Bezalel, Adi Zicher, Yonatan Belinkov, Jonathan Berant, Amir Globerson
Pubblicato in: 2023
Editore: n Chapter of the ACL About Calls Elections Conferences Minutes Officers Policies News RSS Feed The North American Chapter of the Association for Computational Linguistics

Regularizing Towards Permutation Invariance In Recurrent Models

Autori: Cohen, Edo; Ben David, Avichai; Globerson, Amir
Pubblicato in: Advanced in Neural Information Processing Systems (NeurIPS) 2020, 2020
Editore: NeurIPS

On the Inductive Bias of a CNN for Orthogonal Patterns Distributions

Autori: Brutzkus, Alon; Globerson, Amir
Pubblicato in: Uncertainty in Artificial Inteliigence 2021, Numero 1, 2021
Editore: UAI

On the Implicit Bias of Initialization Shape: Beyond Infinitesimal Mirror Descent

Autori: Azulay, Shahar; Moroshko, Edward; Shpigel Nacson, Mor; Woodworth, Blake; Srebro, Nathan; Globerson, Amir; Soudry, Daniel
Pubblicato in: International Conference on Machine Learning 2021, 2021
Editore: ICML

Object-Region Video Transformers

Autori: Roei Herzig, Elad Ben-Avraham, Karttikeya Mangalam, Amir Bar, Gal Chechik, Anna Rohrbach, Trevor Darrell, Amir Globerson
Pubblicato in: 2022
Editore: Conference on Computer Vision and Pattern Recognition (CVPR), 2022
DOI: 10.48550/arxiv.2110.06915

Towards Understanding Learning in Neural Networks with Linear Teachers

Autori: Sarussi, Roy; Brutzkus, Alon; Globerson, Amir
Pubblicato in: International Conference on Machine Learning 2021, 2021
Editore: ICML

PromptonomyViT: Multi-Task Prompt Learning Improves Video Transformers using Synthetic Scene Data

Autori: Roei Herzig, Ofir Abramovich, Elad Ben-Avraham, Assaf Arbelle, Leonid Karlinsky, Ariel Shamir, Trevor Darrell, Amir Globerson
Pubblicato in: 2024
Editore: WACV

In-Context Learning Creates Task Vectors

Autori: Roee Hendel, Mor Geva, Amir Globerson
Pubblicato in: 2023
Editore: EMNLP

EgoPet: Egomotion and Interaction Data from an Animal's Perspective

Autori: Amir Bar, Arya Bakhtiar, Danny Tran, Antonio Loquercio, Jathushan Rajasegaran, Yann LeCun, Amir Globerson, Trevor Darrell
Pubblicato in: European Conference on Computer Vision (ECCV), 2024
Editore: Springer
DOI: 10.48550/arxiv.2404.09991

Stochastic positional embeddings improve masked image modeling

Autori: Amir Bar, Florian Bordes, Assaf Shocher, Mahmoud Assran, Pascal Vincent, Nicolas Ballas, Trevor Darrell, Amir Globerson, Yann LeCun
Pubblicato in: 2024
Editore: ICML

Learning Canonical Representations for Scene Graph to Image Generation

Autori: Herzig, Roei; Bar, Amir; Xu, Huijuan; Chechik, Gal; Darrell, Trevor; Globerson, Amir
Pubblicato in: European Conference on Computer Vision (ECCV, 2020), Numero 2, 2020
Editore: Springer

On the Implicit Bias of Gradient Descent for Temporal Extrapolation

Autori: Cohen-Karlik, Edo; David, Avichai Ben; Cohen, Nadav; Globerson, Amir
Pubblicato in: International Conference on Artificial Intelligence and Statistics (AISTATS), 2022
Editore: PMLR
DOI: 10.48550/arxiv.2202.04302

Implicit Bias of Policy Gradient in Linear Quadratic Control: Extrapolation to Unseen Initial States

Autori: Noam Razin, Yotam Alexander, Edo Cohen-Karlik, Raja Giryes, Amir Globerson, Nadav Cohen
Pubblicato in: 2024
Editore: ICML

Efficient Learning of CNNs using Patch Based Features

Autori: Alon Brutzkus, Amir Globerson, Eran Malach, Alon Regev Netser, Shai Shalev-Schwartz Proceedings
Pubblicato in: 2022
Editore: of the 39th International Conference on Machine Learning, PMLR

Learning to Retrieve Passages without Supervision

Autori: Ori Ram, Gal Shachaf, Omer Levy, Jonathan Berant, Amir Globerson
Pubblicato in: 2021
Editore: arxiv
DOI: 10.48550/arxiv.2112.07708

Bringing Image Structure to Video via Frame-Clip Consistency of Object Tokens

Autori: Elad Ben Avraham, Roei Herzig, Karttikeya Mangalam, Amir Bar, Anna Rohrbach, Leonid Karlinsky , Trevor Darrell, Amir Globerson
Pubblicato in: 2022
Editore: Advanced in Neural Information Processing Systems (NeurIPS)

Incorporating Structured Representations into Pretrained Vision & Language Models Using Scene Graphs

Autori: Roei Herzig, Alon Mendelson, Leonid Karlinsky, Assaf Arbelle, Rogerio Feris, Trevor Darrell, Amir Globerson
Pubblicato in: 2023
Editore: EMNLP

Few-Shot question answering by pretraining span selection

Autori: Ram, Ori; Kirstain, Yuval; Berant, Jonathan; Globerson, Amir; Levy, Omer
Pubblicato in: Association for Computational Linguistics (ACL), 2021, 2021
Editore: ACL-IJCNLP

Differentiable Scene Graphs

Autori: Moshiko Raboh, Roei Herzig, Jonathan Berant, Gal Chechik, Amir Globerson
Pubblicato in: 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 2020, Pagina/e 1477-1486, ISBN 978-1-7281-6553-0
Editore: IEEE
DOI: 10.1109/wacv45572.2020.9093297

On the Inductive Bias of Neural Networks for Learning Read-once DNFs

Autori: Ido Bronstein, Alon Brutzkus, Amir Globerson
Pubblicato in: 2022
Editore: The Conference on Uncertainty in Artificial Intelligence (UAI)

Pre-training Mention Representations in Coreference Models

Autori: Yuval Varkel, Amir Globerson
Pubblicato in: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020, Pagina/e 8534-8540
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.emnlp-main.687

BERTese: Learning to Speak to BERT

Autori: Haviv, Adi; Berant, Jonathan; Globerson, Amir
Pubblicato in: European Chapter of the Association for Computational Linguistics (EACL) 2021, 2021
Editore: EACL

Learning Low Dimensional State Spaces with Overparameterized Recurrent Neural Nets

Autori: Edo Cohen-Karlik, Itamar Menuhin-Gruman, Raja Giryes, Nadav Cohen, Amir Globerson
Pubblicato in: 2023
Editore: The Eleventh International Conference on Learning Representations

Visual Prompting via Image Inpainting

Autori: Amir Bar, Yossi Gandelsman, Trevor Darrell , Amir Globerson, Alexei Efros
Pubblicato in: 2022
Editore: Advanced in Neural Information Processing Systems (NeurIPS)

DETReg: Unsupervised Pretraining with Region Priors for Object Detection

Autori: Bar, Amir; Wang, Xin; Kantorov, Vadim; Reed, Colorado J; Herzig, Roei; Chechik, Gal; Rohrbach, Anna; Darrell, Trevor; Globerson, Amir
Pubblicato in: 2022
Editore: Conference on Computer Vision and Pattern Recognition (CVPR), 2022

Learning Object Permanence from Video

Autori: Shamsian, Aviv; Kleinfeld, Ofri; Globerson, Amir; Chechik, Gal
Pubblicato in: European Conference on Computer Vision (ECCV, 2020), Numero 2, 2020
Editore: Computing Research Repository (CoRR)

Teaching Models to Improve on Tape

Autori: Liat Bezalel, Eyal Orgad, Amir Globerson
Pubblicato in: Proceedings of the AAAI Conference on Artificial Intelligence, Numero 39, 2025, Pagina/e 15550-15558, ISSN 2374-3468
Editore: Association for the Advancement of Artificial Intelligence (AAAI)
DOI: 10.1609/aaai.v39i15.33707

Other (2)

Holdout SGD: Byzantine Tolerant Federated Learning

Autori: Shahar Azulay, Lior Raz, Amir Globerson, Tomer Koren, Yehuda Afek
Pubblicato in: 2020
Editore: arxiv

On the Expressivity of Selective State-Space Layers:A Multivariate Polynomial Approach

Autori: Edo Cohen-Karlik, Itamar Zimerman, Liane Galanti, Ido Atad, Amir Globerson, Lior Wolf
Pubblicato in: 2025
Editore: arxiv
DOI: 10.48550/arxiv.2502.02209