AI Paper Atlas

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

Transfer Learning with Uncertainty Quantification: Random Effect Calibration of Source to Target (RECaST)

Jimmy Hickey, Jonathan P. Williams, Emily C. Hector

JMLR 2024 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Bayesian decision-making under uncertainty · 6 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 RECaST LM instead of target-only DNNDNN · Continuous, n_T=20, sigma^2=0RMSE: 20.5 -> 0.535Improved
use RECaST LM instead of target-only DNNDNN · Continuous, n_T=250, sigma^2=4RMSE: 3.67 -> 14.3Hurt
use RECaST LM instead of Unfreeze DNNUnfreeze DNN · Continuous, n_T=20, sigma^2=0RMSE: 2.49 -> 0.535Improved
use RECaST LM instead of Unfreeze DNNUnfreeze DNN · Continuous, n_T=250, sigma^2=0.25RMSE: 2.76 -> 3.6Hurt
use RECaST DNN instead of target-only DNNDNN · Binary, n_T=20, sigma^2=0AUC: 60.2 -> 97.3Improved
use RECaST DNN instead of target-only DNNDNN · Binary, n_T=250, sigma^2=4AUC: 94.5 -> 89.4Hurt
use RECaST DNN instead of Wiens regularized logistic regressionWiens · Binary, n_T=250, sigma^2=0AUC: 80.4 -> 98.3Improved
use RECaST DNN instead of Unfreeze DNNUnfreeze DNN · Binary, n_T=20, sigma^2=4AUC: 77.1 -> 88.2Improved
use RECaST DNN instead of target-only DNN for prediction setsDNN · Binary coverage, n_T=20, sigma^2=0coverage: 67.2 -> 95.6Improved
increase source-target dissimilarity sigma^2 from 0 to 4RECaST DNN · Binary, n_T=250AUC: 98.3 -> 89.4Hurt
reduce target sample size from 250 to 20RECaST LM · Continuous, sigma^2=0RMSE: 0.517 -> 0.535No clear effect
reduce target sample size from 250 to 20target-only DNN · Continuous, sigma^2=0RMSE: 2.8 -> 20.5Hurt

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