Learning with Confidence
Oliver E. Richardson · 2025
Read Learning with Confidence by Oliver E. Richardson (2025) 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-2508-11037.txt.
Description
People learning from new evidence trust some sources more than others, and Oliver Richardson asks what that trust amounts to in a 2025 paper. His notion of confidence measures how much weight incoming information gets when beliefs update. It looks like probability at a glance, but the paper argues it is a different thing with its own rules.
Richardson lays down axioms for learning with confidence and gives two standard ways to measure it on a scale. He proves confidence can always be put in that form. Under added assumptions, the learning process can be written with vector fields and loss functions, and it supports a language of combined observations. Bayes' rule turns up as a special case: the learner whose loss is a plain average, learning in the most standard way.
Catalog record
| CREATOR | Oliver E. Richardson |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 88 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-11037