Theoretical Understanding of Classic Learning Algorithms

Project Information

TUCLA

Grant agreement ID: 101125203

DOI

10.3030/101125203

EC signature date 15 December 2023

Start date 1 August 2024

End date 31 July 2029

Funded under

• European Research Council (ERC)

Total cost

€ 1 999 288,00

EU contribution

€ 1 999 288,00

Coordinated by

AARHUS UNIVERSITET
Denmark

CORDIS provides links to public deliverables and publications of HORIZON projects.

Links to deliverables and publications from FP7 projects, as well as links to some specific result types such as dataset and software, are dynamically retrieved from OpenAIRE .

Publications

Conference proceedings (20)

Derandomizing Multi-Distribution Learning

Author(s): Kasper Green Larsen, Omar Montasser, Nikita Zhivotovskiy
Published in: Advances in Neural Information Processing Systems 37, 2024
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/079017-2989

Tight Margin-Based Generalization Bounds for Voting Classifiers over Finite Hypothesis Sets

Author(s): Kasper Green Larsen, Natascha Schalburg
Published in: ICML 2026, 2026
Publisher: PMLR
DOI: 10.48550/ARXIV.2511.20407

Revisiting Agnostic Boosting

Author(s): Arthur da Cunha, Mikael Møller Høgsgaard, Andrea Paudice, Yuxin Sun
Published in: Advances in Neural Information Processing Systems 38, 2026
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/085713-4698

Optimal Parallelization of Boosting

Author(s): Arthur Da Cunha, Mikael Møller Høgsgaard, Kasper Green Larsen
Published in: Advances in Neural Information Processing Systems 37, 2024
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/079017-0464

Learning with Monotone Adversarial Corruptions

Author(s): Kasper Green Larsen, Chirag Pabbaraju, Abhishek Shetty
Published in: 2026
Publisher: PMLR

Actively Learning Halfspaces without Synthetic Data

Author(s): Hadley Black, Kasper Green Larsen, Arya Mazumdar, Barna Saha, Geelon So
Published in: 2026
Publisher: PMLR

A New Lower Bound for Multicolor Discrepancy with Applications to Fair Division

Author(s): Ioannis Caragiannis, Kasper Green Larsen, Sudarshan Shyam
Published in: Lecture Notes in Computer Science, Algorithmic Game Theory, 2025
Publisher: Springer Nature Switzerland
DOI: 10.1007/978-3-032-03639-1_13

An Exponential Separation Between Quantum and Quantum-Inspired Classical Algorithms for Linear Systems

Author(s): Allan Grønlund, Kasper Green Larsen
Published in: ICML 2026, 2026
Publisher: PMLR
DOI: 10.48550/ARXIV.2411.02087

The Sample Complexity of Replicable Realizable PAC Learning

Author(s): Kasper Green Larsen, Markus Engelund Mathiasen, Chirag Pabbaraju, Clement Svendsen
Published in: Proceedings of the 58th Annual ACM Symposium on Theory of Computing, 2026
Publisher: ACM
DOI: 10.1145/3798129.3800910

Boosting, Voting Classifiers and Randomized Sample Compression Schemes

Author(s): Arthur da Cunha, Kasper Green Larsen, Martin Ritzert
Published in: 2025
Publisher: PMLR

On Agnostic PAC Learning in the Small Error Regime

Author(s): Julian Asilis, Mikael Møller Høgsgaard, Grigoris Velegkas
Published in: Advances in Neural Information Processing Systems 38, 2026
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/085713-4112

A Fine-Grained Understanding of Uniform Convergence for Halfspaces

Author(s): Aryeh Kontorovich, Kasper Green Larsen
Published in: ICML 2026, 2026
Publisher: PMLR
DOI: 10.48550/ARXIV.2605.06004

Revisiting Agnostic PAC Learning

Author(s): Steve Hanneke, Kasper Green Larsen, Nikita Zhivotovskiy
Published in: 2024 IEEE 65th Annual Symposium on Foundations of Computer Science (FOCS), 2024
Publisher: IEEE
DOI: 10.1109/FOCS61266.2024.00118

Improved Margin Generalization Bounds for Voting Classifiers

Author(s): Mikael Møller Høgsgaard, Kasper Green Larsen
Published in: Proceedings of Thirty Eighth Conference on Learning Theory (series: Proceedings of Machine Learning Research), 2025
Publisher: PMLR

The Interplay Between Interpolation and Aggregation in Regression: Optimal Sample Complexity

Author(s): Mikael Møller Høgsgaard, Kasper Green Larsen, Liang-Yu Zou
Published in: ICML 2026, 2026
Publisher: PMLR
DOI: 10.48550/ARXIV.2605.29819

The Optimal Sample Complexity of Linear Contracts

Author(s): Mikael Møller Høgsgaard
Published in: ICML 2026, 2026
Publisher: PMLR
DOI: 10.48550/ARXIV.2601.01496

The Many Faces of Optimal Weak-to-Strong Learning

Author(s): Mikael Møller Høgsgaard, Kasper Green Larsen, Markus Engelund Mathiasen
Published in: Advances in Neural Information Processing Systems 37, 2024
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/079017-1644

Sample-Near-Optimal Agnostic Boosting with Improved Running Time

Author(s): Arthur da Cunha, Mikael Møller Høgsgaard, Andrea Paudice
Published in: Proceedings of The 37th International Conference on Algorithmic Learning Theory, 2026
Publisher: PMLR

Improved Replicable Boosting with Majority-of-Majorities

Author(s): Kasper Green Larsen, Markus Engelund Mathiasen, Clement Svendsen
Published in: 2026
Publisher: PMLR

Tight Generalization Bounds for Large-Margin Halfspaces

Author(s): Kasper Green Larsen, Natascha Schalburg
Published in: Advances in Neural Information Processing Systems 38, 2026
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)
DOI: 10.52202/085713-1531

Peer reviewed articles (1)

Fourth-moment theorems for sums of multiple integrals

Author(s): Andreas Basse-O’Connor, David Kramer-Bang, Clement Svendsen
Published in: Proceedings of the American Mathematical Society, Issue 154, 2026, ISSN 1088-6826
Publisher: American Mathematical Society (AMS)
DOI: 10.1090/PROC/17808