Training Dynamics in Large Networks: From Super-Wide to the Scaling Law Regime - YouTube
Speaker: Blake Bordelon, Harvard UniversityThis talk is part of the ProbAI Theory of Scaling Laws Workshop 2026, hosted at the Department of Statistics, Univ...
Speaker: Blake Bordelon, Harvard University This talk is part of the ProbAI Theory of Scaling Laws Workshop 2026, hosted at the Department of Statistics, University of Warwick. The workshop is jointly funded by the Centre for Research in Statistical Methodology (CRiSM) at Warwick statistics and Prob_AI Research Hub. Accompanying slides can be found on the workshop website: https://warwick.ac.uk/fac/sci/statistics/news/probai-scaling-laws-2026/programme/ Uh so hello everybody. So uh this morning we have a research talk and a tutorial. So we start with the research talk by Blake. So we met…
saved by
related reading
- Dynamical Mean Field Theory, Random Matrices and Learning in High Dimensionsyoutube.com
- MIA: Greg Yang, Tuning Large Neural Networks via Zero-Shot Hyperparameter Transferyoutube.com
- The Scaling Hypothesis · Gwern.netgwern.net
- Scaling Laws, Carefully | Lil'Loglilianweng.github.io
- Infinite Limits of Neural Networks - Kempner Institutekempnerinstitute.harvard.edu
- On neural scaling and the quanta hypothesisericjmichaud.com
- Scaling is subtler than it seemsberen.io
- [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
- Just Ask for Generalization | Eric Jangevjang.com
- [2205.09653] Self-Consistent Dynamical Field Theory of Kernel Evolution in Wide Neural Networksarxiv.org
- nn-notes.pdfboris-hanin.github.io
- [2608.13335] Neural Quadratic Forms: A Unified Minimal Model for Sudden Learning and Scaling Lawsarxiv.org