// AI news · 2026-11-15 · Nature
Dual Expert ML for Twin Tunnel Settlement Prediction
A source-informed dual expert machine learning model (VF) learns first and incremental settlement from Gaussian and Boltzmann representations, reconstructing cumulative settlement accurately.
Published 2026-11-15 08:00 (Beijing time)Source NatureAuto-summarised, fact-checked against source
01Summary
In nested five-fold validation of 52 staged HSsmall scenarios, VF achieves R² = 0.839 and RMSE = 0.2315 cm for ΔS, matching the Gaussian expert and outperforming the Boltzmann expert. VF outperforms conventional baselines in both S1 and S2 prediction, with lowest ΔS RMSE among tested methods. Extrapolation performance degrades when complete response families are absent from training.
Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.
02Key points
- VF learns S1 and ΔS from Gaussian and Boltzmann representations
- ΔS prediction RMSE of 0.2315 cm outperforms Boltzmann expert
- VF achieves lowest ΔS error among baseline models
- Extrapolation fails when full response families are missing
03Source
Nature · published 2026-11-15 08:00 Beijing time
05Related news
- Same source · Explainable IoT Intrusion Detection via Bio-Inspired Encoding2026-11-15 · Nature
- Same source · HGS-BioAbductor: Explainable Survival Modeling and Hypothesis Abduction2026-11-14 · Nature
- Same topic · Gemini 4 Carbon wins MathArena benchmark2026-11-15 · The Decoder
- Same topic · Google launches Teamwork system to verify mathematical proofs2026-11-14 · IT之家
- Same day · Tech Execs Petition for Nvidia to Reinstatement of Jensen Huang2026-11-15 · TechCrunch AI
From AIMap AI news: collected from 1,100+ official and media sources. Summaries only restate the original article; follow the link for the full story. All stories of 2026-11-15 · Archive by date · All AI news · Today