On improving generalization in a class of learning problems with the method of small parameters for weakly-controlled optimal gradient systems
Getachew K. Befekadu · 2024
Read On improving generalization in a class of learning problems with the method of small parameters for weakly-controlled optimal gradient systems by Getachew K. Befekadu (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-08772.txt.
Description
Getachew Befekadu, a solo author, builds a math frame to improve learning from limited data in a 2024 paper. The problem class covers point estimates for high-dimensional nonlinear functions. A gradient system steered by a weak control drives the model, and training data plus a noisy copy guide its motion over time.
Small-parameter methods split the hard problem into a chain of simpler ones. Solving each in turn and pooling the answers gives estimates that generalize better, the author argues. A rate bound tells how fast the pooled answer nears the true optimum. Tests on a standard nonlinear regression case round out the work.
Catalog record
| CREATOR | Getachew K. Befekadu |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-08772