Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.22809 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908614115983360 |
|---|---|
| author | Hazard, Christopher J. Resnick, Michael Beel, Jacob Xia, Jack Mack, Cade Glennie, Dominic Fulp, Matthew Maze, David Bassett, Andrew Koistinen, Martin |
| author_facet | Hazard, Christopher J. Resnick, Michael Beel, Jacob Xia, Jack Mack, Cade Glennie, Dominic Fulp, Matthew Maze, David Bassett, Andrew Koistinen, Martin |
| contents | Traditional machine learning relies on explicit models and domain assumptions, limiting flexibility and interpretability. We introduce a model-free framework using surprisal (information theoretic uncertainty) to directly analyze and perform inferences from raw data, eliminating distribution modeling, reducing bias, and enabling efficient updates including direct edits and deletion of training data. By quantifying relevance through uncertainty, the approach enables generalizable inference across tasks including generative inference, causal discovery, anomaly detection, and time series forecasting. It emphasizes traceability, interpretability, and data-driven decision making, offering a unified, human-understandable framework for machine learning, and achieves at or near state-of-the-art performance across most common machine learning tasks. The mathematical foundations create a ``physics'' of information, which enable these techniques to apply effectively to a wide variety of complex data types, including missing data. Empirical results indicate that this may be a viable alternative path to neural networks with regard to scalable machine learning and artificial intelligence that can maintain human understandability of the underlying mechanics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_22809 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Theory of the Mechanics of Information: Generalization Through Measurement of Uncertainty (Learning is Measuring) Hazard, Christopher J. Resnick, Michael Beel, Jacob Xia, Jack Mack, Cade Glennie, Dominic Fulp, Matthew Maze, David Bassett, Andrew Koistinen, Martin Machine Learning Artificial Intelligence Statistics Theory Traditional machine learning relies on explicit models and domain assumptions, limiting flexibility and interpretability. We introduce a model-free framework using surprisal (information theoretic uncertainty) to directly analyze and perform inferences from raw data, eliminating distribution modeling, reducing bias, and enabling efficient updates including direct edits and deletion of training data. By quantifying relevance through uncertainty, the approach enables generalizable inference across tasks including generative inference, causal discovery, anomaly detection, and time series forecasting. It emphasizes traceability, interpretability, and data-driven decision making, offering a unified, human-understandable framework for machine learning, and achieves at or near state-of-the-art performance across most common machine learning tasks. The mathematical foundations create a ``physics'' of information, which enable these techniques to apply effectively to a wide variety of complex data types, including missing data. Empirical results indicate that this may be a viable alternative path to neural networks with regard to scalable machine learning and artificial intelligence that can maintain human understandability of the underlying mechanics. |
| title | A Theory of the Mechanics of Information: Generalization Through Measurement of Uncertainty (Learning is Measuring) |
| topic | Machine Learning Artificial Intelligence Statistics Theory |
| url | https://arxiv.org/abs/2510.22809 |