My Publications

You can also find all my publications on Google Scholar.

Preprints

★ means equal contribution, sometimes in arbitrary order

1. On the Necessity of Adaptive Regularisation: Optimal Anytime Online Learning on $\boldsymbol{\ell_p}$-Balls
   Emmeran Johnson, D. Martínez-Rubio, Ciara Pike-Burke, Patrick Rebeschini (2025)
   [PDF] [BibTeX]📋 copy

   @article{johnson2025optimal,
     title={Optimal anytime online learning on $\ell_p$-balls},
     author={Johnson, Emmeran and Mart{\'\i}nez-Rubio, David and Ciara Pike-Burke and Rebeschini, Patrick},
     journal={arXiv preprint arXiv:2501.19752},
     url={https://arxiv.org/abs/2501.19752},
     year={2025}
   }
   

2. Beyond Short Steps in Frank-Wolfe Algorithms
   D. Martínez-Rubio★, S. Pokutta★ (2025)
   [PDF] [BibTeX]📋 copy

   @article{martinezrubio2025beyond,
     title={Beyond Short Steps in Frank-Wolfe Algorithms},
     author={Mart{\'\i}nez-Rubio, David and Pokutta, Sebastian},
     journal={arXiv preprint arXiv:2501.18773},
     url={https://arxiv.org/abs/2501.18773},
     year={2025}
   }
   

3. Linear Convergence of the Frank-Wolfe Algorithm over Product Polytopes
   G. Iommazzo, D. Martínez-Rubio, P. Criado, E. Wirth, S. Pokutta (2025)
   [PDF] [BibTeX]📋 copy

   @article{iommazzo2025linear,
     title={Linear Convergence of the Frank-Wolfe Algorithm over Product Polytopes},
     author={Iommazzo, Gabriele and Mart{\'\i}nez-Rubio, David and Criado, Francisco and Wirth, Elias and  Pokutta, Sebastian},
     journal={arXiv preprint arXiv:2505.11259},
     url={https://arxiv.org/abs/2505.11259},
     year={2025}
   }
   

4. Strong Convexity of Sets in Riemannian Manifolds
   D. Scieur★, D. Martínez-Rubio★, T. Kerdreux★, A. d'Aspremont, S. Pokutta (2023)
   [PDF] [BibTeX]📋 copy

   @article{scieur2023strong,
     title={Strong Convexity of Sets in Riemannian Manifolds},
     author={Scieur, Damien and Kerdreux, Thomas  and Mart{\'\i}nez-Rubio, David and d'Aspremont, Alexandre and Pokutta, Sebastian},
     journal={arXiv preprint arXiv:2312.03583},
     url={https://arxiv.org/abs/2312.03583},
     year={2023}
   }
   

Conference Proceedings

1. Non-Euclidean High-Order Smooth Convex Optimization
   J.P. Contreras★, C. Guzmán★, D. Martínez-Rubio★ (2024)
   COLT 2025, 36th Annual Conference on Learning Theory
   [PDF] [BibTeX]📋 copy

   @article{contreras2024noneuclidean,
     title={Non-Euclidean High-Order Smooth Convex Optimization},
     author={Contreras, Juan Pablo and Guzmán, Cristóbal and Mart{\'\i}nez-Rubio, David},
     journal={arXiv preprint arXiv:2411.08987},
     url={https://arxiv.org/abs/2411.08987},
     year={2024}
   }
   

2. Implicit Riemannian Optimism with Applications to Min-Max Problems
   C. Roux★, D. Martínez-Rubio★, S. Pokutta (2025)
   ICML 2025, 42nd International Conference on Machine Learning
   [PDF] [BibTeX]📋 copy

   @article{roux2025implicit,
     title={Implicit Riemannian Optimism with Applications to Min-Max Problems},
     author={Roux, Christophe and Mart{\'\i}nez-Rubio, David and Pokutta, Sebastian},
     journal={arXiv preprint arXiv:2501.18381},
     url={https://arxiv.org/abs/2501.18381},
     year={2025}
   }
   

3. Secant Line Search for Frank-Wolfe Algorithms
   D. Hendrych★, S. Pokutta★, M. Besançon★, D. Martínez-Rubio★, (2025)
   ICML 2025, 42nd International Conference on Machine Learning
   [PDF] [BibTeX]📋 copy

   @article{hendrych2025secant,
     title={Secant Line Search for Frank-Wolfe Algorithms},
     author={Hendrych, Deborah and Pokutta, Sebastian and Besançon, Mathieu and Mart{\'\i}nez-Rubio, David},
     journal={arXiv preprint arXiv:2501.18775},
     url={https://arxiv.org/abs/2501.18775},
     year={2025}
   }
   

4. Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties
   D. Martínez-Rubio★, C. Roux★, C. Criscitiello, S. Pokutta (2023)
   AISTATS 2025, 28th International Conference on Artificial Intelligence and Statistics
   [PDF] [BibTeX]📋 copy

   @article{martinez2023accelerated,
     title={Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties},
     author={Mart{\'\i}nez-Rubio, David and Roux, Christophe and Criscitiello, Christopher and Pokutta, Sebastian},
     journal={arXiv preprint arXiv:2305.16186},
     url={https://arxiv.org/abs/2305.16186},
     year={2023}
   }
   

5. Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization
   S. Vary, D. Martínez-Rubio, Patrick Rebeschini (2024)
   AISTATS 2025, 28th International Conference on Artificial Intelligence and Statistics
   [PDF] [BibTeX]📋 copy

   @article{vary2024blackbox,
     title={Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization},
     author={Vary, Simon and Mart{\'\i}nez-Rubio, David and Rebeschini, Patrick},
     journal={arXiv preprint arXiv:2412.15956},
     url={https://arxiv.org/abs/2412.15956},
     year={2024}
   }
   

6. Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian Proximal Point
   D. Martínez-Rubio★, C. Roux★, S. Pokutta
   ICML 2024, 41th International Conference on Machine Learning
   [PDF] [BibTeX]📋 copy

   @inproceedings{martinez2024convergence,
     author       = {David Mart{\'{\i}}nez{-}Rubio and
                     Christophe Roux and
                     Sebastian Pokutta},
     title     = {Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian
                     Proximal Point},
     booktitle = {Proceedings of the 41st International Conference on Machine Learning, {ICML} 2024
                     Vienna, Austria, July 21-27},
     series    = {Proceedings of Machine Learning Research},
     volume    = {235},
     pages     = {34920-34948},
     publisher = { {PMLR} },
     year      = {2024},
     url       = {https://arxiv.org/abs/2403.10429},
   }
   

7. Accelerated and Sparse Algorithms for Approximate Personalized PageRank and Beyond
   D. Martínez-Rubio★, E. Wirth★, and S. Pokutta
   COLT 2023, 36th Annual Conference on Learning Theory
   [PDF] [Slides] [BibTeX]📋 copy

   @inproceedings{martinezrubio2023pagerank,
     author       = {David Mart{\'{\i}}nez{-}Rubio and
                     Elias Samuel Wirth and
                     Sebastian Pokutta},
     editor       = {Gergely Neu and
                     Lorenzo Rosasco},
     title        = {Accelerated and Sparse Algorithms for Approximate Personalized PageRank
                     and Beyond},
     booktitle    = {The Thirty Sixth Annual Conference on Learning Theory, {COLT} 2023,
                     12-15 July 2023, Bangalore, India},
     series       = {Proceedings of Machine Learning Research},
     volume       = {195},
     pages        = {2852--2876},
     publisher    = { {PMLR} },
     year         = {2023},
     url          = {https://arxiv.org/abs/2303.12875},
   }
   

8. Accelerated Riemannian Optimization: Handling Constraints with a Prox to Bound Geometric Penalties
   D. Martínez-Rubio, and S. Pokutta
   COLT 2023, 36th Annual Conference on Learning Theory
   [PDF] [Slides] [BibTeX]📋 copy

   @inproceedings{martinezRubio2022accelerated,
     author       = {David Mart{\'{\i}}nez{-}Rubio and
                     Sebastian Pokutta},
     editor       = {Gergely Neu and
                     Lorenzo Rosasco},
     title        = {Accelerated Riemannian Optimization: Handling Constraints with a Prox
                     to Bound Geometric Penalties},
     booktitle    = {The Thirty Sixth Annual Conference on Learning Theory, {COLT} 2023,
                     12-15 July 2023, Bangalore, India},
     series       = {Proceedings of Machine Learning Research},
     volume       = {195},
     pages        = {359--393},
     publisher    = { {PMLR} },
     year         = {2023},
     url          = {https://arxiv.org/abs/2211.14645},
   }
   

9. Fast Algorithms for Packing Proportional Fairness and its Dual
   F. Criado★, D. Martínez-Rubio★, S. Pokutta
   NeurIPS 2022, 35th Advances in Neural Information Processing Systems
   [PDF] [Slides] [BibTeX]📋 copy

   @inproceedings{criado2022fast,
     title={Fast algorithms for packing proportional fairness and its dual},
     author={Criado, Francisco and Mart{\'i}nez-Rubio, David and Pokutta, Sebastian},
     booktitle    = {Advances in Neural Information Processing Systems 35: Annual Conference
                     on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans,
                     LA, USA, November 28 - December 9, 2022},
     volume={35},
     pages={25754--25766},
     url={https://arxiv.org/abs/2109.03678},
     year={2022}
   }
   

10. Global Riemannian Acceleration in Hyperbolic and Spherical Spaces
    D. Martínez-Rubio
    ALT 2022, 33rd International Conference on Algorithmic Learning Theory
    [PDF] [Slides] [BibTeX]📋 copy

    @inproceedings{martinezRubio2020global,
      author    = {David Mart{\'{i}}nez{-}Rubio},
      editor    = {Sanjoy Dasgupta and
                   Nika Haghtalab},
      title     = {Global Riemannian Acceleration in Hyperbolic and Spherical Spaces},
      booktitle = {International Conference on Algorithmic Learning Theory, 29-1 April
                   2022, Paris, France},
      series    = {Proceedings of Machine Learning Research},
      volume    = {167},
      pages     = {768--826},
      publisher = { {PMLR} },
      year      = {2022},
      url       = {https://proceedings.mlr.press/v167/martinez-rubio22a.html},
    }
    
    
    

11. Decentralized Cooperative Stochastic Bandits
    D. Martínez-Rubio, V. Kanade, and P. Rebeschini
    NeurIPS 2019, 32nd Advances in Neural Information Processing Systems
    [PDF] [Slides] [BibTeX]📋 copy [Code]

    @inproceedings{martinezRubio2018decentralized,
      author    = {David Mart{\'{i}}nez{-}Rubio and
                   Varun Kanade and
                   Patrick Rebeschini},
      editor    = {Hanna M. Wallach and
                   Hugo Larochelle and
                   Alina Beygelzimer and
                   Florence d'Alch{\'{e}}{-}Buc and
                   Emily B. Fox and
                   Roman Garnett},
      title     = {Decentralized Cooperative Stochastic Bandits},
      booktitle = {Advances in Neural Information Processing Systems 32: Annual Conference
                   on Neural Information Processing Systems 2019, NeurIPS 2019, December
                   8-14, 2019, Vancouver, BC, Canada},
      pages     = {4531--4542},
      year      = {2019},
      url       = {https://proceedings.neurips.cc/paper/2019/hash/85353d3b2f39b9c9b5ee3576578c04b7-Abstract.html},
    }
    

12. Cheap Orthogonal Constraints in Neural Networks: A Simple Parametrization of the Orthogonal and Unitary Group
    M. Lezcano-Casado★, and D. Martínez-Rubio★
    ICML 2019, 36th International Conference on Machine Learning
    [PDF] [Slides] [BibTeX]📋 copy

    @inproceedings{DBLP:conf/icml/CasadoM19,
      author    = {Mario Lezcano Casado and
                   David Mart{\'{i}}nez{-}Rubio},
      editor    = {Kamalika Chaudhuri and
                   Ruslan Salakhutdinov},
      title     = {Cheap Orthogonal Constraints in Neural Networks: {A} Simple Parametrization
                   of the Orthogonal and Unitary Group},
      booktitle = {Proceedings of the 36th International Conference on Machine Learning,
                   {ICML} 2019, 9-15 June 2019, Long Beach, California, {USA}},
      series    = {Proceedings of Machine Learning Research},
      volume    = {97},
      pages     = {3794--3803},
      publisher = { {PMLR} },
      year      = {2019},
      url       = {http://proceedings.mlr.press/v97/lezcano-casado19a.html},
    }
    

13. Online Learning Rate Adaptation with Hypergradient Descent
    A. G. Baydin, R. Cornish, D. Martínez-Rubio, M. Schmidt, and F. Wood
    ICLR 2018, 6th International Conference on Learning Representations
    [PDF] [BibTeX]📋 copy

    @inproceedings{DBLP:conf/iclr/BaydinCMSW18,
      author    = {Atilim Gunes Baydin and
                   Robert Cornish and
                   David Mart{\'{i}}nez{-}Rubio and
                   Mark Schmidt and
                   Frank Wood},
      title     = {Online Learning Rate Adaptation with Hypergradient Descent},
      booktitle = {6th International Conference on Learning Representations, {ICLR} 2018,
                   Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedings},
      publisher = {OpenReview.net},
      year      = {2018},
      url       = {https://openreview.net/forum?id=BkrsAzWAb},
    }
    

Workshop Publications and Technical Reports

• Open Problem: Polynomial linearly-convergent method for g-convex optimization?
  C. Criscitiello★, D. Martínez-Rubio★, and N. Boumal★
  COLT 2023, 36th Annual Conference on Learning Theory
  [PDF] [BibTeX]📋 copy

  @inproceedings{criscitiello2023polynomial,
    author       = {Christopher Criscitiello and
                    David Mart{\'{\i}}nez{-}Rubio and
                    Nicolas Boumal},
    editor       = {Gergely Neu and
                    Lorenzo Rosasco},
    title        = {Open Problem: Polynomial linearly-convergent method for g-convex optimization?},
    booktitle    = {The Thirty Sixth Annual Conference on Learning Theory, {COLT} 2023,
                    12-15 July 2023, Bangalore, India},
    series       = {Proceedings of Machine Learning Research},
    volume       = {195},
    pages        = {5950--5956},
    publisher    = { {PMLR} },
    year         = {2023},
    url          = {https://proceedings.mlr.press/v195/criscitiello23b.html},
  }
  

• Neural Networks are a priori Biased Towards Boolean Functions with Low Entropy
  C. Mingard, J. Skalse, G. Valle-Pérez, D. Martínez-Rubio, V. Mikulik, and A. A. Louis (2019)
  [PDF] [BibTeX]📋 copy

  @article{mingard2019neural,
    title={Neural networks are a priori biased towards Boolean functions with low entropy},
    author={Mingard, Chris and Skalse, Joar and Valle-P{\'e}rez, Guillermo and Mart{\'i}nez-Rubio, David and Mikulik, Vladimir and Louis, Ard A},
    journal={arXiv preprint arXiv:1909.11522},
    url={http://arxiv.org/pdf/1909.11522.pdf},
    year={2019}
  }
  

• Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators
  M. Lezcano-Casado, A.G. Baydin, D. Martínez-Rubio, T.A. Le, F. Wood, L. Heinrich, G. Louppe, K. Cranmer, K. Ng, W. Bhimji and Prabhat
  The Workshop on Deep Learning for Physical Sciences, NIPS 2017
  [PDF] [BibTeX]📋 copy

  @article{DBLP:journals/corr/abs-1712-07901,
    author    = {Mario Lezcano Casado and
                 Atilim Gunes Baydin and
                 David Mart{\'{i}}nez{-}Rubio and
                 Tuan Anh Le and
                 Frank D. Wood and
                 Lukas Heinrich and
                 Gilles Louppe and
                 Kyle Cranmer and
                 Karen Ng and
                 Wahid Bhimji and
                 Prabhat},
    title     = {Improvements to Inference Compilation for Probabilistic Programming
                 in Large-Scale Scientific Simulators},
    journal   = {CoRR},
    volume    = {abs/1712.07901},
    year      = {2017},
    url       = {http://arxiv.org/pdf/1712.07901.pdf},
    archivePrefix = {arXiv},
    eprint    = {1712.07901},
  }
  
  

Theses

• PhD Thesis
  Acceleration in First-Order Optimization Methods: Promenading Beyond Convexity or Smoothness, and Applications
  University of Oxford, 2021
  [PDF] [BibTeX]📋 copy

   
  @article{martinezrubio2021acceleration,
    title={Acceleration in First-Order Optimization Methods: Promenading Beyond Convexity or Smoothness, and Applications},
    author={Mart{\'i}nez-Rubio, David},
    journal={University of Oxford, Oxford, PhD thesis},
    url={https://damaru2.github.io/phd_thesis/thesis.pdf},
    year={2021}
  }

• MSc Thesis
  Convergence Analysis of an Adaptive Method of Gradient Descent
  University of Oxford, 2017
  [PDF] [BibTeX]📋 copy

  @article{martinezrubio2017convergence,
    title={Convergence analysis of an adaptive method of gradient descent},
    author={Mart{\'i}nez-Rubio, David},
    journal={University of Oxford, Oxford, M. Sc. thesis},
    url={https://damaru2.github.io/convergence_analysis_hypergradient_descent/dissertation_hypergradients.pdf},
    year={2017}
  }

• BSc Thesis
  Characterization and computation of the Galois group of polynomials of degree 7
  Complutense University of Madrid, 2016
  [PDF] [BibTeX]📋 copy

  @article{martinezrubio2016characterization,
    title={Characterization and computation of the Galois group of polynomials of degree 7},
    author={Mart{\'i}nez-Rubio, David},
    journal={Complutense University of Madrid, Madrid, B.Sc. Thesis},
    url={https://damaru2.github.io/assets/TFG.pdf},
    year={2016}
    }