Computational Learning Theory: compact representation, efficient computation, and societal challenges in learning MDPs

Informazioni relative al progetto

COLT-MDP

ID dell’accordo di sovvenzione: 882396

DOI

10.3030/882396

Data della firma CE 29 Aprile 2020

Data di avvio 1 Ottobre 2020

Data di completamento 30 Settembre 2026

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale

€ 1 878 125,00

Contributo UE

€ 1 878 125,00

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 (49)

Agnostic Reinforcement Learning with Low-Rank MDPs and Rich Observations

Autori: Ayush Sekhari, Christoph Dann, Mehryar Mohri, Yishay Mansour, Karthik Sridharan
Pubblicato in: NeurIPS 2021b, 2021
Editore: Conference on Neural Information Processing Systems

A Constant Approximation Algorithm for Sequential Random-Order No-Substitution k-Median Clustering

Autori: Tom Hess, Michal Moshkovitz, Sivan Sabato
Pubblicato in: NeurIPS 2021g, 2021
Editore: Conference on Neural Information Processing Systems

Principal-Agent Reward Shaping in MDPs

Autori: Omer Ben-Porat, Yishay Mansour, Michal Moshkovitz, Boaz Taitler
Pubblicato in: AAAI 2024, 2024
Editore: AAAI 2024

Efficient Rate Optimal Regret for Adversarial Contextual MDPs Using Online Function Approximation

Autori: Orin Levy, Alon Cohen, Asaf B. Cassel, Yishay Mansour
Pubblicato in: ICML 2023c, 2023
Editore: PMLR

Near-Optimal Regret for Adversarial MDP with Delayed Bandit Feedback

Autori: Tiancheng Jin, Tal Lancewicki, Haipeng Luo, Yishay Mansour, Aviv Rosenberg
Pubblicato in: NeurIPS 2022b, 2022
Editore: NeurIPS 2022, Arxiv, European Workshop on Reinforcement Learning (EWRL 2022), Complex Feedback in Online Learning (CFOL ), 2022

Learnability Gaps of Strategic Classification

Autori: Lee Cohen, Yishay Mansour, Shay Moran, Han Shao
Pubblicato in: COLT 2024, 2024
Editore: COLT 2024

Stochastic Strategic Patient Buyers: Revenue maximization using posted prices

Autori: Eitan-Hai Mashiah, Idan Attias, Yishay Mansour
Pubblicato in: AAAI 2023b, 2023
Editore: AAAI

Learning Adversarial Markov Decision Processes with Delayed Feedback

Autori: Tal Lancewicki, Aviv Rosenberg, Yishay Mansour
Pubblicato in: AAAI 2022b, 2022
Editore: the Association for the Advancement of Artificial Intelligence ; Cambridge University Press

Oracle-Efficient Reinforcement Learning in Factored MDPs with Unknown Structure

Autori: Aviv Rosenberg, Yishay Mansour
Pubblicato in: NeurIPS 2021h, 2021
Editore: NeurIPS

Non-stochastic Bandits With Evolving Observations

Autori: Yogev Bar-On, Yishay Mansour
Pubblicato in: ALT 2025, 2025
Editore: ALT 2025

A Learning-Theoretic Framework for Certified Auditing of Machine Learning Models

Autori: Chhavi Yadav, Michal Moshkovitz, Kamalika Chaudhuri
Pubblicato in: NeurIPS 2022c, 2022
Editore: NeurIPS 2022, Arxiv, European Workshop on Reinforcement Learning (EWRL 2022), Complex Feedback in Online Learning (CFOL ), 2022

Benign Underfitting of Stochastic Gradient Descent

Autori: Tomer Koren, Roi Livni, Yishay Mansour, Uri Sherman
Pubblicato in: NeurIPS 2022a, 2022
Editore: NeurIPS 2022

A Theory of Interpretable Approximations

Autori: Marco Bressan, Nicolò Cesa-Bianchi, Emmanuel Esposito, Yishay Mansour, Shay Moran, Maximilian Thiessen
Pubblicato in: COLT 2024, 2024
Editore: COLT 2024

Prediction with Corrupted Expert Advice

Autori: Idan Amir, Idan Attias, Tomer Koren, Yishay Mansour, Roi Livni
Pubblicato in: NeurIPS 2020a, 2020
Editore: NeurIPS 2020

Eluder-based Regret for Stochastic Contextual MDPs

Autori: Orin Levy, Asaf Cassel, Alon Cohen, Yishay Mansour
Pubblicato in: FC 2024, 2024
Editore: FC 2024

Oracle-Efficient Regret Minimization in Factored MDPs with Unknown Structure

Autori: Aviv Rosenberg, Yishay Mansour
Pubblicato in: NeurIPS 2021e, 2021
Editore: Conference on Neural Information Processing Systems

Improved Regret for Efficient Online Reinforcement Learning with Linear Function Approximation

Autori: Uri Sherman, Tomer Koren, Yishay Mansour:
Pubblicato in: ICML 2023d, 2023
Editore: PMLR

A Characterization of Semi-Supervised Adversarially-Robust PAC Learnability

Autori: Idan Attias, Steve Hanneke, Yishay Mansour
Pubblicato in: NeurIPS 2022d, 2022
Editore: NeurIPS

Stochastic Shortest Path with Adversarially Changing Costs

Autori: Aviv Rosenberg, Yishay Mansour
Pubblicato in: IJCAI 2021, 2021
Editore: IJCAI

Optimal Rates for Random Order Online Optimization

Autori: Uri Sherman, Tomer Koren, Yishay Mansour
Pubblicato in: NeurIPS 2021d, 2021
Editore: Conference on Neural Information Processing Systems

he Real Price of Bandit Information in Multiclass Classification

Autori: Liad Erez, Alon Cohen, Tomer Koren, Yishay Mansour, Shay Moran
Pubblicato in: COLT 2024, 2024
Editore: COLT 2024

Regret Minimization and Convergence to Equilibria in General-sum Markov Games

Autori: Liad Erez, Tal Lancewicki, Uri Sherman, Tomer Koren, Yishay Mansour
Pubblicato in: ICML 2023b, 2023
Editore: PMLR

Dynamic algorithms against an adaptive adversary: generic constructions and lower bounds.

Autori: Amos Beimel, Haim Kaplan, Yishay Mansour, Kobbi Nissim, Thatchaphol Saranurak, Uri Stemmer
Pubblicato in: STOC 2022, 2022
Editore: STOC 2022

Reinforcement Learning Can Be More Efficient with Multiple Rewards

Autori: Christoph Dann, Yishay Mansour, Mehryar Mohri
Pubblicato in: ICML 2023a, 2023
Editore: International Conference on Machine Learning

Data Reconstruction: When You See It and When You Don't

Autori: Edith Cohen, Haim Kaplan, Yishay Mansour, Shay Moran, Kobbi Nissim, Uri Stemmer, Eliad Tsfadia
Pubblicato in: ITCS 2025, 2025
Editore: ITCS 2025

Learning Efficiently Function Approximation for Contextual MDP

Autori: Orin Levy, Yishay Mansour
Pubblicato in: EWRL 2022, 2022
Editore: European Workshop on Reinforcement Learning

Sample Complexity of Uniform Convergence for Multicalibration

Autori: Eliran Shabat, Lee Cohen, Yishay Mansour
Pubblicato in: NeurIPS 2020c, 2020
Editore: Conference on Neural Information Processing Systems

FriendlyCore: Practical Differentially Private Aggregation

Autori: Eliad Tsfadia, Edith Cohen, Haim Kaplan, Yishay Mansour, Uri Stemmer
Pubblicato in: ICML 2022d, 2022
Editore: ICML 2022

Modeling Attrition in Recommender Systems with Departing Bandits

Autori: Omer Ben-Porat, Lee Cohen, Liu Leqi, Zachary C. Lipton, Yishay Mansour
Pubblicato in: AAAI 2022a, 2022
Editore: the Association for the Advancement of Artificial Intelligence

Faster Convergence with MultiWay Preferences.

Autori: Aadirupa Saha, Vitaly Feldman, Yishay Mansour, Tomer Koren
Pubblicato in: AISTATS 2024, 2024, Pagina/e 433-441
Editore: AISTATS 2024

Partially Interpretable Models with Guarantees on Coverage and Accuracy

Autori: Nave Frost, Zachary C. Lipton, Yishay Mansour, Michal Moshkovitz:
Pubblicato in: ALT2024, 2024
Editore: ALT 2024

Stochastic Multi-Armed Bandits with Unrestricted Delay Distributions

Autori: Tal Lancewicki, Shahar Segal, Tomer Koren, Yishay Mansour
Pubblicato in: ICML 2021g, 2021
Editore: Cambridge University Press

Dueling Bandits with Team Comparisons

Autori: Lee Cohen, Yishay Mansour, Ulrike Schmidt-Kraepelin
Pubblicato in: NeurIPS 2021c, 2021
Editore: Conference on Neural Information Processing Systems

Separating Adaptive Streaming from Oblivious Streaming

Autori: Haim Kaplan, Yishay Mansour, Kobbi Nissim, Uri Stemmer
Pubblicato in: CRYPTO 2021, 2021
Editore: The International Association for Cryptologic Research

Private Learning of Halfspaces: Simplifying the Construction and Reducing the Sample Complexity

Autori: Haim Kaplan, Yishay Mansour, Uri Stemmer, Eliad Tsfadia
Pubblicato in: NeurIPS 2020b, 2020
Editore: Conference on Neural Information Processing Systems

Differentially Private Multi-Armed Bandits in the Shuffle Model

Autori: Jay Tenenbaum, Haim Kaplan, Yishay Mansour, Uri Stemmer
Pubblicato in: NeurIPS 2021a, 2021
Editore: Conference on Neural Information Processing Systems

Learning-Augmented Algorithms with Explicit Predictors

Autori: Marek Eliás, Haim Kaplan, Yishay Mansour, Shay Moran
Pubblicato in: NeurIPS 2024, 2024
Editore: NeurIPS 2024

Minimax Regret for Stochastic Shortest Path

Autori: Alon Cohen, Yonathan Efroni, Yishay Mansour, Aviv Rosenberg
Pubblicato in: NeurIPS 2021f, 2021
Editore: Conference on Neural Information Processing Systems

Differentially-Private Clustering of Easy Instances

Autori: Edith Cohen, Haim Kaplan, Yishay Mansour, Uri Stemmer, Eliad Tsfadia
Pubblicato in: ICML 2021f, 2021
Editore: International Conference on Machine Learning 2021 (ICML)

Guarantees for Epsilon-Greedy Reinforcement Learning with Function Approximation

Autori: Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
Pubblicato in: ICML 2022b, 2022
Editore: International Conference on Machine Learning (ICML 2022)

Rate-Optimal Policy Optimization for Linear Markov Decision Processes

Autori: Uri Sherman, Alon Cohen, Tomer Koren, Yishay Mansour
Pubblicato in: ICML 2024, 2024
Editore: ICML 2024

Optimism in Face of a Context: Regret Guratees for Stochastic Contextual MDP

Autori: Orin Levy, Yishay Mansour
Pubblicato in: AAAI 2023a, 2022
Editore: AAAI 2023, European Workshop on Reinforcement Learning (EWRL 2022), Complex Feedback in Online Learning (CFOL), 2022

Fair Wrapping for Black-box Predictions

Autori: Alexander Soen, Ibrahim Alabdulmohsin, Sanmi Koyejo, Yishay Mansour, Nyalleng Moorosi, Richard Nock, Ke Sun, Lexing Xie
Pubblicato in: NeurIPS 2022e, 2022
Editore: NeurIPS

Optimal Publishing Strategies on a Base Layer

Autori: Yogev Bar-On, Yishay Mansour
Pubblicato in: FC 2024, 2024
Editore: FC2024

Adversarial Dueling Bandits

Autori: Aadirupa Saha, Tomer Koren, Yishay Mansour
Pubblicato in: ICML 2021a, 2021
Editore: Cambridge University Press

Cooperative Online Learning in Stochastic and Adversarial MDPs

Autori: Tal Lancewicki, Aviv Rosenberg, Yishay Mansour
Pubblicato in: ICML 2022c, 2022
Editore: International Conference on Machine Learning (ICML 2022), European Workshop on Reinforcement Learning (EWRL 2022)

Fast Rates for Bandit PAC Multiclass Classification

Autori: Liad Erez, Alon Cohen, Tomer Koren, Yishay Mansour, Shay Moran
Pubblicato in: NeurIPS 2024, Numero 2024, 2024
Editore: NeurIPS 2024

The Sparse Vector Technique, Revisited

Autori: Haim Kaplan, Yishay Mansour, Uri Stemmer
Pubblicato in: COLT 2021, 2021
Editore: Association for Computational Learning

Dueling Convex Optimization

Autori: Aadirupa Saha, Tomer Koren, Yishay Mansour
Pubblicato in: ICML 2021e, 2021
Editore: International Conference on Machine Learning

Book chapters (2)

Uniswap Liquidity Provision: An Online Learning Approach

Autori: Yogev Bar-On, Yishay Mansour
Pubblicato in: Lecture Notes in Computer Science, Financial Cryptography and Data Security. FC 2023 International Workshops, 2023, Pagina/e 247-261
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-48806-1_16

A Theory of Learning with Competing Objectives and User Feedback

Autori: Pranjal Awasthi, Corinna Cortes, Yishay Mansour, Mehryar Mohri
Pubblicato in: Lecture Notes in Computer Science, Artificial Intelligence and Image Analysis, 2024, Pagina/e 10-49
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-63735-3_2

Peer reviewed articles (2)

Adversarially Robust Streaming Algorithms via Differential Privacy

Autori: Avinatan Hassidim, Haim Kaplan, Yishay Mansour, Yossi Matias, Uri Stemmer:
Pubblicato in: JACM, NeurIPS 2020d, Numero Journal of the ACM 69(6): 42:1-42:14 2022,, 2022, ISSN 0004-5411
Editore: Association for Computing Machinary, Inc.

A machine-learning-based alternative to phylogenetic bootstrap

Autori: Noa Ecker , Dorothée Huchon , Yishay Mansour , Itay Mayrose , Tal Pupko
Pubblicato in: Bioinformatics, Volume 40, Numero 1, July 2024, 2024, Pagina/e Pages i208–i217,, ISSN 1367-4803
Editore: Oxford University Press

Other (3)

Finding Safe Zones of Markov Decision Processes Policies

Autori: Lee Cohen, Yishay Mansour, Michal Moshkovitz
Pubblicato in: 2022
Editore: Archive

There is no Accuracy-Interpretability Tradeoff in Reinforcement Learning for Mazes

Autori: Yishay Mansour, Michal Moshkovitz, Cynthia Rudin
Pubblicato in: 2022
Editore: Arxiv

Differentially-Private Bayes Consistency

Autori: Olivier Bousquet, Haim Kaplan, Aryeh Kontorovich, Yishay Mansour, Shay Moran, Menachem Sadigurschi, Uri Stemmer
Pubblicato in: 2022
Editore: Archive