Neural Bregman Divergences For Distance Learning

Abstract

Many metric learning tasks, such as triplet learning, nearest neighbor retrieval, and visualization, are treated primarily as embedding tasks where the ultimate metric is some variant of the Euclidean distance (e.g., cosine or Mahalanobis), and the algorithm must learn to embed points into the pre-chosen space. The study of non-Euclidean geometries is often not explored, which we believe is due to a lack of tools for learning non-Euclidean measures of distance. Recent work has shown that Bregman divergences can be learned from data, opening a promising approach to learning asymmetric distances. We propose a new approach to learning arbitrary Bergman divergences in a differentiable manner via input convex neural networks and show that it overcomes significant limitations of previous works. We also demonstrate that our method more faithfully learns divergences over a set of both new and previously studied tasks, including asymmetric regression, ranking, and clustering. Our tests further

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylu2022neural

Related papers

• Metric Learning On Manifolds (2019)0.00
• Learning Distances From Data With Normalizing Flows And Score Matching (2024)0.00
• Divide And Conquer The Embedding Space For Metric Learning (2019)14.39
• Variance & Greediness: A Comparative Study Of Metric-learning Losses (2026)0.00
• No Fuss Distance Metric Learning Using Proxies (2017)20.03
• Discriminative Learning Of Similarity And Group Equivariant Representations (2018)0.00
• Directional Statistics-based Deep Metric Learning For Image Classification And Retrieval (2018)13.05
• Learning Deep Optimal Embeddings With Sinkhorn Divergences (2022)0.00