Roi Holtzman
0 followers · 315 views
on the atlas — 7
- Marcus' Blog4 savers
- Fit Without Fear: An Over-Fitting Perspective on Modern Deep and Shallow Learning - YouTube1 savers
- Articles – Epoch AI1 savers
- LessWrong39 savers
- The necessity of machine learning theory in mitigating AI risk3 savers
- Infinite Limits of Neural Networks - Kempner Institute3 savers
- Curius / Onboarding2621 savers
highlights — 3
Furthermore, the fact that neural architectures are often successful across multiple application domains, suggests that their effectiveness relies on fundamental patterns in data, a “gravitational force” in the data solar system, rather than a serendipitous “alignment of the planets” in specific instances of data analysis. This universality hints at a continuity with other fundamental principles discovered in science and mathematics.
The necessity of machine learning theory in mitigating AI riskOne then scales the width up and sees that those hyperparameters remain nearly optimal for the wider networks.
Infinite Limits of Neural Networks - Kempner InstituteHe went over to his grandfather
Curius / Onboarding