AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Sparse Bayesian Deep Functional Learning with Structured Region Selection

Xiaoxian Zhu, Yingmeng Li, Shuangge Ma, Mengyun Wu

ICML 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Bayesian view of meta-learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use sBayFDNN instead of linear region selectors (BFRS, SLoS, cFuSIM) on ECGsBayFDNN · ECGregion F1: 0.412 -> 0.634 (vs SLoS)Improved
use sBayFDNN instead of baselines on TecatorsBayFDNN · TecatorRMSE: 2.217 -> 2.138 (vs FNN)Improved
use sBayFDNN instead of baselines on Bike rentalsBayFDNN · Bike rentalRMSE: 0.684 -> 0.618 (vs SLoS)Improved
add structured first-layer sparsity versus plain spline-feature FNNspline-feature DNN · ECGRMSE: 12.991 -> 12.069Improved
increase spline bases Jn from small to moderate (80) (ablation)sBayFDNN · Simulation (simple beta, composite g, SNR 10)region F1: 0.5573 -> 0.9323Improved
increase spline bases Jn to very large (240) (ablation)sBayFDNN · Simulation (simple beta, composite g, SNR 10)region F1: 0.9323 -> 0.7583Hurt
raise PIP threshold from 0.5 to 0.9 (ablation)sBayFDNN · Simulation macro-averageregion F1: 0.8781 -> 0.4557Hurt
lower PIP threshold from 0.5 to 0.3 (ablation)sBayFDNN · Simulation macro-averageregion F1: 0.8781 -> 0.8969Improved
shrink spike variance to 1e-6 (ablation)sBayFDNN · Simulation macro-averageregion F1: 0.7982 -> 0.5463Hurt
vary prior inclusion probability lambda_n (ablation)sBayFDNN · Simulation macro-averageregion F1: 0.7868 to 0.8182 across valuesNo clear effect
use evidence-based selection of JnsBayFDNN · Simulation (composite g, SNR 10)within 30% of oracle F1: 97.1%Improved
increase observation grid Tobs from 200 to 2000 (ablation)sBayFDNN · Simulation runtime (n=1000)runtime per repeat (s): 125.1 -> 103.7No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy