Accelerating Generative Models and Nonconvex Optimisation

I have organised (together with Sotirios Sabanis) a two-week Theory and Methods Challenges Fortnight (TMCF) event on Accelerating generative models and nonconvex optimisation. The events took place in person and online at Turing HQ based in the British Library, London. The first week took place from June 6th to 10th, 2022, and the second week took place from September 5th to 9th, 2022. The final workshop of the event occurred on March 24th, 2023, at Turing. Recordings of this workshop are now available.

The event consisted of several challenges and participants tackling them. It has been a lot of fun -- if you are interested in organising such an event, check TMCF calls from Turing webpage.

We list below the outcomes of this event, i.e., works that are initiated during the event.

  • On diffusion-based generative models and their error bounds: The log-concave case with full convergence estimates, S. Bruno, Y. Zhang, D. Lim, O. D. Akyildiz, S. Sabanis, 2023, [arXiv].
  • Tweedie Moment Projected Diffusions For Inverse Problems, Benjamin Boys, Mark Girolami, Jakiw Pidstrigach, Sebastian Reich, Alan Mosca, O. Deniz Akyildiz, 2023, [arXiv].
  • Interacting Particle Langevin Algorithm for Maximum Marginal Likelihood Estimation, Ö. D. Akyildiz, F. R. Crucinio, M. Girolami, T. Johnston, S. Sabanis, 2023, [arXiv].
  • Random Grid Neural Processes for Parametric Partial Differential Equations, A. Vadeboncoeur, I. Kazlauskaite, F. Cirak, M. Girolami, Ö. D. Akyildiz, International Conference of Machine Learning (ICML), 2023, [arXiv].
  • Convergence of denoising diffusion models under the manifold hypothesis, V. De Bortoli, 2022, Transactions on Machine Learning Research, [journal].
  • Score-based generative Models Introduction, J. Pidstrigach, 2022, [link].
  • DiffusionJAX, B. Boys (Lead), 2023, [Github Repo].