Explaining hidden mechanisms: a generative model for causal graphs with nonlinear latent factors
Koji Maruhashi, Heewon Park, Rui Yamaguchi, Satoru Miyano · 2026
Read Explaining hidden mechanisms: a generative model for causal graphs with nonlinear latent factors by Koji Maruhashi, Heewon Park, Rui Yamaguchi, Satoru Miyano (2026) 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/pmc-13219030.txt.
Description
Koji Maruhashi, Heewon Park, Rui Yamaguchi, and Satoru Miyano tackled a hard problem in two thousand twenty-six: cause-and-effect graphs from medical data are often too tangled to read. Their model assumes each graph comes from a small set of hidden causal drivers. A rule keeps the hidden and observed effects in line with each other.
Their method, called TRIP, learns those shared hidden drivers from graphs and condition labels. It balances a prediction loss with a graph rebuild loss, so the result stays both useful and easy to read. Tests on fake data and on cancer cell gene networks show it finds hidden patterns and improves prediction.
Catalog record
| CREATOR | Koji Maruhashi, Heewon Park, Rui Yamaguchi, Satoru Miyano |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |