Gradient Span Algorithms Make Predictable Progress in High Dimension
We prove that all 'gradient span algorithms' have asymptotically deterministic behavior on scaled Gaussian random functions as the dimension tends to infinity. In particular, this result explains the counterintuitive phenomenon that different training runs of many large machine learning models result in approximately equal cost curves despite random initialization on a complicated non-convex landscape. The distributional assumption of (non-stationary) isotropic Gaussian random functions we use is sufficiently general to serve as realistic model for machine learning training but also encompass spin glasses and random quadratic functions.
Oct-13-2024
- Country:
- North America
- United States
- Ohio (0.04)
- New York > New York County
- New York City (0.14)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- California > San Diego County
- San Diego (0.04)
- Arizona > Maricopa County
- Phoenix (0.04)
- Canada
- Quebec > Montreal (0.04)
- British Columbia > Metro Vancouver Regional District
- Vancouver (0.04)
- United States
- Europe
- Asia
- Africa > Middle East
- Tunisia > Ben Arous Governorate > Ben Arous (0.04)
- North America
- Genre:
- Research Report (0.64)
- Technology: