Nonstationary Distance Metric Learning

Abstract

Recent work in distance metric learning has focused on learning transformations of data that best align with provided sets of pairwise similarity and dissimilarity constraints. The learned transformations lead to improved retrieval, classification, and clustering algorithms due to the better adapted distance or similarity measures. Here, we introduce the problem of learning these transformations when the underlying constraint generation process is nonstationary. This nonstationarity can be due to changes in either the ground-truth clustering used to generate constraints or changes to the feature subspaces in which the class structure is apparent. We propose and evaluate COMID-SADL, an adaptive, online approach for learning and tracking optimal metrics as they change over time that is highly robust to a variety of nonstationary behaviors in the changing metric. We demonstrate COMID-SADL on both real and synthetic data sets and show significant performance improvements relative to previo

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keygreenewald2016nonstationary

Related papers

• Dynamic Metric Learning From Pairwise Comparisons (2016)3.58
• Similarity Function Tracking Using Pairwise Comparisons (2017)2.26
• Large Scale Local Online Similarity/distance Learning Framework Based On Passive/aggressive (2018)2.26
• Sparse Online Relative Similarity Learning (2021)2.26
• Discrete Scale-invariant Metric Learning For Efficient Collaborative Filtering (2025)0.00
• No Fuss Distance Metric Learning Using Proxies (2017)20.03
• A New Similarity Space Tailored For Supervised Deep Metric Learning (2020)3.58
• Adaptive Neighborhood Metric Learning (2022)10.07