AI Paper Atlas

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

Variational Bayesian Unlearning

Quoc Phong Nguyen, Bryan Kian Hsiang Low, Patrick Jaillet

NeurIPS 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Variational Bayesian deep learning · 44 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
minimize reverse KL instead of EUBO (ablation)Bayesian unlearning without adjusted likelihood (lambda=0) · Synthetic linear regressionKL divergence: 266.68 -> 0.0158Improved
add adjusted likelihood (lambda 0 -> 0.1) to EUBO (ablation)EUBO unlearning · Synthetic linear regressionKL divergence: 266.68 -> 0.0899Improved
add adjusted likelihood to reverse KL (lambda 0 -> 0.1) (ablation)rKL unlearning · Synthetic linear regressionKL divergence: 0.0158 -> 0.0600Hurt
add adjusted likelihood to EUBO (lambda 0 -> 1e-9) (ablation)EUBO unlearning · Moon sparse GP classificationavg predictive KLImproved
add adjusted likelihood to EUBO (lambda 0 -> 1e-20) (ablation)EUBO unlearning · Airline sparse GPKL divergence: 2629.71 -> 1384.96Improved
add adjusted likelihood to rKL (lambda 0 -> 1e-13) (ablation)rKL unlearning · Airline sparse GPKL divergence: 455.11 -> 367.12Improved
reverse KL instead of EUBO (ablation)unlearning · Airline sparse GPKL divergence: 2629.71 -> 455.11Improved
replace Gaussian posterior with normalizing flow (ablation)EUBO and rKL on logistic regression · Banknote authenticationavg predictive KLImproved
unlearning with rKL at lambda=0 vs no unlearning (ablation)Fashion MNIST logistic regression · Fashion MNISTavg predictive KLImproved
EUBO at lambda=0 vs adjusted lambda (ablation)EUBO unlearning · Fashion MNISTavg predictive KLHurt

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