AI Paper Atlas

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

Efficient Variational Bayes Learning of Graphical Models With Smooth Structural Changes

Hang Yu, Songwei Wu, Justin Dauwels

TPAMI 2022 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Variational Bayesian deep learning · 7 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 BASS instead of cross-validated KERNELgraphical lasso with kernel smoothing · Synthetic (P=500)F1-score: 0.38 -> 0.72Improved
use BASS instead of oracle-tuned KERNELKERNEL · Synthetic (P=500)F1-score: 0.58 -> 0.72Improved
add temporal smoothness (LOGGLE) over KERNEL under oracle tuningKERNEL · Synthetic (P=100)F1-score: 0.75 -> 0.80Improved
select tuning parameters by CV instead of oracleLOGGLE · Synthetic (P=100)F1-score: 0.80 -> 0.59Hurt
select tuning parameters by CV instead of oracleSINGLE · Synthetic (P=100)F1-score: 0.79 -> 0.58Hurt
increase sample size N from 500 to 2000BASS · Synthetic (P=20)F1-score: 0.77 -> 0.92Improved
increase graph density Ne from 10 to 40BASS · Synthetic (P=20, N=1000)F1-score: 0.92 -> 0.74Hurt
increase dimension P from 20 to 500BASS · Synthetic (N=1000)F1-score: 0.88 -> 0.72Hurt
use BASS instead of oracle GMS in frequency domainGMS · Synthetic time series (P=20)F1-score: 0.76 -> 0.89Improved
use BASS instead of GMS (CV) for EEG graphsGMS (CV) · EEG MCIclassification accuracy: 65.85% -> 80.00%Improved
use BASS instead of GMS (CV) for EEG graphsGMS (CV) · EEG mild ADclassification accuracy: 70% -> 85.37%Improved
use BASS instead of GMS (CV) for edge-count group testGMS (CV) · EEG MCIMann-Whitney p-value: 3.99e-1 -> 7.55e-3Improved

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