VaRDASS improves unsupervised domain adaptation by using stratified sampling to reduce variance in discrepancy estimation for measures like correlation alignment and MMD, with derived error bounds, an optimality proof for MMD under assumptions, and a k-means style algorithm.
SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives
2 Pith papers cite this work, alongside 445 external citations. Polarity classification is still indexing.
2
Pith papers citing it
445
external citations · Pith
abstract
In this work we introduce a new optimisation method called SAGA in the spirit of SAG, SDCA, MISO and SVRG, a set of recently proposed incremental gradient algorithms with fast linear convergence rates. SAGA improves on the theory behind SAG and SVRG, with better theoretical convergence rates, and has support for composite objectives where a proximal operator is used on the regulariser. Unlike SDCA, SAGA supports non-strongly convex problems directly, and is adaptive to any inherent strong convexity of the problem. We give experimental results showing the effectiveness of our method.
fields
cs.LG 2verdicts
UNVERDICTED 2representative citing papers
Year-wise cross-validation across ten ML algorithms on Harmonized Landsat-Sentinel imagery shows SVMs achieve mean F1 of 0.74 for almonds in California and 0.59 for corn in Iowa by early June in unseen validation years.
citing papers explorer
-
Variance Matters: Improving Domain Adaptation via Stratified Sampling
VaRDASS improves unsupervised domain adaptation by using stratified sampling to reduce variance in discrepancy estimation for measures like correlation alignment and MMD, with derived error bounds, an optimality proof for MMD under assumptions, and a k-means style algorithm.
-
Intercomparison of Machine Learning Algorithms for Remote Sensing-based In-season Crop Mapping
Year-wise cross-validation across ten ML algorithms on Harmonized Landsat-Sentinel imagery shows SVMs achieve mean F1 of 0.74 for almonds in California and 0.59 for corn in Iowa by early June in unseen validation years.