GeoERM learns orthonormal task representations via Riemannian gradient descent with polar retraction, and reports accuracy gains over Euclidean multi-task baselines.
(2021), Predicting with proxies: Transfer learning in high dimension, Management Science\/ , 67, 2964--2984
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GeoERM: Geometry-Aware Multi-Task Representation Learning on Riemannian Manifolds
GeoERM learns orthonormal task representations via Riemannian gradient descent with polar retraction, and reports accuracy gains over Euclidean multi-task baselines.