Metric Learning: A Survey

By Brian Kulis. 2013

999 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

In Representation & Self-Supervised Learning

Citations in the corpus

Dual Contradistinctive Generative Autoencoder (2020, positive): “…the above observations, we develop a new generative autoencoder model with dual contradistinctive losses by adopting a discriminative loss performing instance-level classiﬁcation (enforcing the instance-level ﬁdelity), which is rooted in metric learning [40] and contrastive learning [20, 65, 62].”
Prototypical Networks for Few-shot Learning (2017, methodology): “The literature on metric learning is vast [17, 5]; we summarize here the work most relevant to our proposed method.”
Learning Visual Clothing Style with Heterogeneous Dyadic Co-Occurrences (2015, background): “Then, our goal is to learn a function r = retrieve(Iq, j) to retrieve an item r such that comp(q, r) and q ∈ Ci, r ∈ Cj , i 6= j.”
Deep Weighted Averaging Classifiers (2018, methodology): “This is closely related to a long line of past work on metric learning [3, 10, 23, 58, 60], but rather than trying to optimize a particular notion of distance (such as Mahalanobis distance), we make use of a fixed distance function, and allow the architecture of standard deep models to do the equiva”
Few-Shot Learning with Metric-Agnostic Conditional Embeddings (2018, background): “Our work draws on a number of previous approaches to few-shot learning, predominantly those referred to under the umbrella of metric-learning (Kulis et al., 2013; Vinyals et al., 2016; Snell et al., 2017).”
Learning Metrics From Teachers: Compact Networks for Image Embedding (2019, background): “There is a large number of works on metric learning, see for example the survey [16].”
A Survey on Metric Learning for Feature Vectors and Structured Data (2013, methodology): “A more recent survey, written independently and in parallel to our work, is due to Kulis (2012). Despite some overlap, it should be noted that both surveys have their own strengths and complement each other well.”
Task-Embedded Control Networks for Few-Shot Imitation Learning (2018, background): “The success of our new approach comes from learning a metric space, and there has been an abundance of work in metric learning for image classification [23, 24], from which we will summarise the most relevant.”
General Heterogeneous Transfer Distance Metric Learning via Knowledge Fragments Transfer (2017, methodology): “The framework is based on a generalized notion of the Mahalanobis distance [Kulis, 2012].”
Collaborative Metric Learning (2017, background): “Mathematically, a metric needs to satisfy several conditions, among which the triangle inequality is the most cru- c©2017 International World Wide Web Conference Committee (IW3C2), published under Creative Commons CC BY 4.0 License.”
Metric Learning for Novelty and Anomaly Detection (2018, background): “Deriving such a measure from data samples is known as metric learning [14].”
Fine-grained visual categorization via multi-stage metric learning (2014, background): “Many algorithms have been developed for DML [11, 32, 33] and a detailed review can be found in two survey papers [19, 34].”
Towards Integrative Machine Learning and Knowledge Extraction (2017, methodology): “Thus, RNNs are widely used in applications such as speech recognition or in activity recognition [58], whereas in image processing mainly feed-forward networks are of relevance [59].”
Deep Metric Learning with Angular Loss (2017, neutral): “[19] on early works of metric learning.”
Becoming the expert - interactive multi-class machine teaching (2015, methodology): “This gives us the ﬂex-ibility of allowing similarity to be deﬁned using feature vectors extracted from the images, human provided attributes, or using distance metric learning [23].”
Subspace Robust Wasserstein distances (2019, methodology): “The optimal transport (OT) toolbox (Villani, 2009) is gaining popularity in machine learning, with several applications to data science outlined in the recent review paper (Peyré & Cuturi, 2019).”
Self-Supervised Learning in Remote Sensing: A review (2022, background): “Therefore, more attention should be attributed to the encoder architecture and a negative sampling strategy related to metric learning [173].”