AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Multi-view Collaborative Gaussian Process Dynamical Systems

Shiliang Sun, Jingjing Fei, Jing Zhao, Liang Mao

JMLR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Uncertainty-aware multi-view learning

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
replace the GPLVM prior on private latents with the mixture-mean/covariance GP prior (ablation)McGPDS · Human motion (silhouette to pose)RMSE: 2.62± 0.04 -> 2.37± 0.03Improved
replace the GPLVM prior on private latents with the mixture-mean/covariance GP prior (ablation)McGPDS · Human motion (silhouette to pose)MSLL: 3.78±0.25 -> 2.60±0.05Improved
use CGPDS output coupling layer instead of a linear map (ablation)McGPDS · Human motion (silhouette to pose)RMSE: 2.81± 0.15 -> 2.37± 0.03Improved
share latent space across views instead of independent CGPDS per view (ablation)McGPDS / MRD with dynamics · Human motion (silhouette to pose)RMSE: 2.69±0.13 -> 2.37± 0.03Improved
add dynamics to MRD (ablation)MRD · Human motion (silhouette to pose)RMSE: 5.03±0.01 -> 2.65±0.01Improved
McGPDS instead of MRD with dynamicsmulti-view GPDS · CUAVE (video frames to MFCC)MSLL: 3.45±0.20 -> 1.94±0.07Improved
use latent X learned by McGPDS instead of MRD for nearest neighbourNNXspace · CUAVE (video frames to MFCC)RMSE: 1.70±0.10 -> 1.38±0.15Improved
McGPDS instead of nearest neighbour on small non-dynamic classification dataMcGPDS · Oil dataset (classification)accuracyHurt
increase private latent dimension / inducing points (ablation)McGPDS · Human motion (silhouette to pose)RMSEImproved

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