Dr. Stefano Bruno
I am a Visiting Postdoctoral Researcher at the University of Edinburgh (UK) and a Postdoctoral Research Associate at UNIST (Korea).
My research lies at the interface of Machine Learning, Optimization, and Probability. I develop theoretically grounded stochastic optimization and sampling algorithms for nonconvex and high-dimensional problems, with a particular focus on first- and zeroth-order optimization, Langevin dynamics, and generative modeling.
Previously, I was a Postdoctoral Research Associate in the School of Mathematics at the University of Edinburgh (UK). During my time in Edinburgh, I secured a £500,000 UK–South Korea Collaborative R&D grant, funded by Innovate UK, to support my research in collaboration with Neurolabs. The project has also raised an additional £415,000 in South Korea, funded by the KIAT (Korea Institute for Advancement of Technology, 한국산업기술진흥원).
From November 2024 to April 2025, I worked as a Senior Researcher at Neurolabs, a UK-based Computer Vision startup.
I completed my PhD in Mathematics at the University of Bath (UK), under the supervision of Prof. Hendrik Weber, as part of the Probability Laboratory (Prob-L@B) and the EPSRC Centre for Doctoral Training in Statistical Applied Mathematics. My doctoral research was in Stochastic Analysis, with particular emphasis on Stochastic Partial Differential Equations.
Offices:
301-2, 112 Building, 50 UNIST-gil, Ulsan, 44919, Republic of Korea
4610, JCMB, King's Building Campus, University of Edinburgh, EH9 3FD, UK
Publications
Wasserstein Convergence of Score-based Generative Models under Semiconvexity and Discontinuous Gradients
TMLR: Transactions on Machine Learning Research.
On diffusion-based generative models and their error bounds: The log-concave case with full convergence estimates
TMLR: Transactions on Machine Learning Research.
Optimal regularity in time and space for stochastic porous medium equations
Annals of Probability, 50(6), 2288–2343.
Note for ML:
Annals of Probability is a top-tier journal in Probability Theory.
Research Pipeline
Submitted Papers Under Review
- One manuscript under review at a top-tier machine learning conference (title to be announced after the review process).
- One manuscript under review at a leading optimization journal (title to be announced in November).
Work in Preparation
- Four manuscripts currently in preparation on optimization for machine learning and quantitative finance.
Reviewer
- NeurIPS: Conference on Neural Information Processing Systems. Top Reviewer Award (NeurIPS 2026, Top 10%).
- ICML: International Conference on Machine Learning.
- ICLR: International Conference on Learning Representations.
- JMLR: Journal of Machine Learning Research.
- JASA: Journal of the American Statistical Association.
- TMLR: Transactions on Machine Learning Research.
Grants
-
Innovate UK: circa 500,000 GBP.
Additional funding (over 415,000 GBP) is provided by
KIAT (Korea Institute for Advancement of Technology,
한국산업기술진흥원) as part of the UK-South Korea
collaborative R&D grant.
Project: AI innovation in the supply chain of consumer packaged-goods for recognising objects in retail execution, supply chain management and smart factories: using novel diffusion-based optimization algorithms and diffusion-based generative models.
A brief description of the project is available on the UKRI website. Further information is provided on the announcement by the School of Mathematics at the University of Edinburgh and the Neurolabs press release.
- London Mathematical Society: 3,000 GBP & UoE School of Mathematics Knowledge Exchange fund: 1,000 GBP. Support for organising the Hackathon Workshop on Generative Modeling.
Hackathon on Generative AI
I was the lead organiser of the Hackathon Workshop on Generative Modeling during the Isaac Newton Institute–Alan Turing Institute satellite programme on Diffusions in machine learning: Foundations, generative models and non-convex optimization. Two industry-inspired challenges were provided by Amazon (Edinburgh).
Collaborators
- Ömer Deniz Akyıldız (Imperial College London, UK)
- Benjamin Gess (TU Berlin & Max Planck Institute, Germany)
- Dong-Young Lim (UNIST, Korea)
- Sotirios Sabanis (University of Edinburgh, UK)
- Hendrik Weber (University of Bath, PhD supervisor now in Münster)
- Ying Zhang (HKUST (GZ), China)
PhD Defense Examiners: Prof. Franco Flandoli (SNS Pisa) and Prof. Alexander Cox (University of Bath).
Talks, Research Visits, and Conferences
Talks
- Presentation: "AI-Guided Discovery of Perovskite Passivation Molecules", AI-Space Solar Initiative InnoCORE Workshop, Busan, South Korea, 20th–21st August 2026.
- Presentation: "Flatness-Aware Stochastic Gradient Langevin Dynamics", 43rd International Conference on Machine Learning (ICML 2026), Seoul, South Korea, 7th–9th July 2026.
- Special session on "Stochastic Computation and Complexity", 17th International Conference on Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing (MCQMC), University of Edinburgh, UK, 8th–12th June 2026.
- Tech seminar, Lagrange Mathematics and Computing Research Center, Paris, France, 21st February 2024.
- Probability seminar, University of Manchester, UK, 29th November 2023.
- North British Probability Workshop, University of Edinburgh, UK, 3rd May 2023.
- BICMR Analysis and PDE seminar, Peking University, China, 16th December 2021.
- Bielefeld Stochastic Afternoon, Bielefeld University, Germany, 15th December 2021.
- Prob-L@B internal monthly seminar, University of Bath, UK, 13th December 2021.
- SAMBa Conference 2021, University of Bath, UK, 5th–7th July 2021.
- Welcome Home Unito 2020, University of Turin, Italy, 21st–22nd December 2020.
Research Visits
- Scientific programme in Diffusions in machine learning: Foundations, generative models and non-convex optimisation at the Isaac Newton Institute/Alan Turing Institute, London, UK. Invited Participant and Organiser, 24th June–12th July 2024.
- Invited by Prof. Ying Zhang at the Hong Kong University of Science and Technology (Guangzhou), China, May 2024.
- Machine Learning and Finance Laboratory of Prof. Dong-Young Lim at Ulsan National Institute of Science and Technology (UNIST) in Ulsan, Republic of Korea, May 2023 and May 2024.
- Stochastic Partial Differential Equations group of Prof. Benjamin Gess at the Max-Planck Institute for Mathematical Sciences (MIS) in Leipzig, Germany, December 2019.
- Scientific programme in Scaling limits, rough paths, quantum field theory at the Isaac Newton Institute for Mathematical Sciences, University of Cambridge, UK. Invited Participant, 26th–30th November and 10th–14th December 2018.
Conferences and Research Schools
- 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026), Virtual Pass (Top Reviewer), 8th–10th December 2026.
- AI-Space Solar Initiative InnoCORE Workshop, Busan, South Korea, 21st–23rd December 2025.
- Summer school on fundamentals and applications of diffusion models, Isaac Newton Institute for Mathematical Sciences satellite programme, UK, 24th–28th June 2024.
- 11th International Conference on Stochastic Analysis and its Applications, International Centre for Mathematical Sciences (ICMS), Edinburgh, UK, 26th–30th June 2023.
- Accelerating Generative Models and Nonconvex Optimisation, Alan Turing Institute, London, UK, 23rd March 2023.
- Pathwise Stochastic Analysis and Applications, CIRM, Luminy, France, 8th–12th March 2021.
- 41st Conference on Stochastic Processes and their Applications, Northwestern University, Evanston, USA, 8th–12th July 2019.
- Paths between Probability, PDEs, and Physics, Imperial College London, 1st–5th July 2019.
- Workshop on the Theory and Applications of Stochastic Partial Differential Equations, Fields Institute, Toronto, Canada, 10th–14th June 2019.
- New Directions in Stochastic Analysis: Rough Paths, SPDEs and Related Topics, Weierstrass Institute, Berlin, Germany, 18th–22nd March 2019.
- Spring School on Random Interfaces, University of Augsburg, Germany, 13th–15th March 2019.
- BMS Summer School: Probabilistic and statistical methods for networks, TU Berlin, Germany, 21st August–1st September 2017.
- 26th Jyväskylä Summer School: Lévy and Feller Processes and Applications, University of Jyväskylä, Finland, 8th–12th August 2016.
Teaching
Tutorials (University of Bath, UK)
- Probability 2A. Semester 1, Academic years: 2020/2021, 2021/2022.
- Vector Calculus and PDEs. Semester 2, Academic year: 2019/2020.
- Analysis 2A. Semester 1, Academic year: 2019/2020.
- Analysis 1. Semesters 1–2, Academic years: 2019/2020, 2020/2021.
- Analysis of nonlinear ODEs. Semester 2, Academic year: 2020/2021.
- Mathematical methods 2. Semester 2, Academic year: 2020/2021.