Adrián Javaloy

prof-pic.jpg

I am a Lecturer in AI and Probabilistic Machine Learning at the University of Bath, as well as a member of the Centre for Artificial Intelligence and ELLIS.

Before that, I was a postdoc at the april lab working with Antonio Vergari and a PhD student at the Probabilistic Machine Learning group advised by Isabel Valera.

My research focuses on developing methods that are reliable, principled, and efficient. My ultimate goal is to conduct interesting and principled science to better understand machine learning models and their post-training behaviour.

A non-exhaustive list of my current interests could be:

  • Probabilistic machine learning.
  • Tractable probabilistic models.
  • Causal generative models.
  • Constrained optimisation.

I am actively looking for PhD students and visitors!

If interested, send me an email explaining how our research interests align and we can discuss options.

news

Sep 02, 2026 🙈 I have joined the University of Bath as a Lecturer!
Jul 28, 2026 🗣️ I gave a talk at the Oberwolfach workshop on Geometric methods in optimization
Jul 14, 2026 🧑🏻‍🏫 I taught with Robert Peharz a course on Tractable Circuits at ESSAI 2026
May 07, 2026 🎉 An Embarrassingly Simple Way to Optimize Orthogonal Matrices at Scale will be at ICML 2026
Jan 25, 2026 🎉 How to Square Tensor Networks and Circuits Without Squaring Them will be at ICLR 2026

selected publications

  1. An Embarrasingly Simple Way to Optimize Orthogonal Matrices at Scale
    Adrián Javaloy and Antonio Vergari
    In Forty-third International Conference on Machine Learning, 2026
  2. How to Square Tensor Networks and Circuits Without Squaring Them
    Lorenzo Loconte, Adrián Javaloy, and Antonio Vergari
    In The Fourteenth International Conference on Learning Representations, ICLR 2026, 2026
  3. Preprint
    COPA: Comparing the incomparable in multi-objective model evaluation
    Adrián Javaloy, Antonio Vergari, and Isabel Valera
    arXiv preprint arXiv:2503.14321, 2025
  4. DeCaFlow: A deconfounding causal generative model
    Alejandro Almodóvar*, Adrián Javaloy*, Juan Parras, and 2 more authors
    In The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025
  5. Dissertation
    Meet my expectations: on the interplay of trustworthiness and deep learning optimization
    Adrián Javaloy
    2024
  6. Causal normalizing flows: from theory to practice
    Adrián Javaloy, Pablo Sánchez-Martín, and Isabel Valera
    In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023, 2023