Improved Deep Metric Learning with Multi-class N-pair Loss Objective

By Kihyuk Sohn. 2016

2,481 citations recorded by the source corpus.

Subfields: 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 Representation & Self-Supervised Learning

Citations in the corpus

Learning Transferable Visual Models From Natural Language Supervision (2021, background): “…knowledge this batch construction technique and objective was first introduced in the area of deep metric learning as the multi-class N-pair loss Sohn (2016), was popularized for contrastive representation learning by Oord et al. (2018) as the InfoNCE loss, and was recently adapted for…”
A Simple Framework for Contrastive Learning of Visual Representations (2020, methodology): “This loss has been used in previous work (Sohn, 2016; Wu et al., 2018; Oord et al., 2018); for convenience, we term it NT-Xent (the normalized temperature-scaled cross entropy loss).”
Contrastive Multiview Coding (2019, methodology): “[26, 81, 72, 25, 31, 57, 3, 29, 36] that take advantage of the co-occurence of multiple views.”
Unsupervised Feature Learning via Non-parametric Instance Discrimination (2018, background): “Metric learning has been shown to be effective for few-shot learning [38, 41, 37].”
Improving Text-to-Image Synthesis Using Contrastive Learning (2021, methodology): “We utilize the Normalized Temperature-scaled Cross Entropy Loss (NT-Xent) [2, 30, 41] as the contrastive loss.”
On Mutual Information Maximization for Representation Learning (2019, background): “In the particular case that x and y take value in the same space and f is constrained to be of the form f ( x, y ) = φ ( x ) (cid:62) φ ( y ) , for some function φ , this coincides (up to constants and change of sign) with the expectation of the multi-class K-pair loss proposed in (Sohn, 2016, Eqn.”
Unsupervised Embedding Learning via Invariant and Spreading Instance Feature (2019, methodology): “The pre-trained Inception-V1 [39] on ImageNet is used as the backbone network following existing methods [30, 32, 37].”
Unified Contrastive Learning in Image-Text-Label Space (2022, methodology): “With this goal, numerous works have pushed the image recognition performance from different directions, such as data scale from MNIST [36] to ImageNet1K [10], model architectures from convolutional neural networks (CNNs) [23, 26, 34, 35, 40, 52, 53] to Transformers [15, 43, 57, 61, 64, 67, 72], and ”
What makes for good views for contrastive learning (2020, background): “2 Related Work Recently the most competitive methods for learning representations without labels have been selfsupervised contrastive representation learning [42, 26, 58, 53, 49, 7].”
Time-Contrastive Networks: Self-Supervised Learning from Video (2017, neutral): “We also experiment with other metric learning losses, namely npairs [42] and lifted structured [43], and show that results are comparable.”
Learning Spread-Out Local Feature Descriptors (2017, methodology): “In the experiment section, we test combining the global orthogonal regularization with contractive loss3 (1) [19], triplet loss (2) [1], lifted structured similarity softmax loss (LSSS) [14] and N-pair loss [21].”
Supervised Contrastive Learning (2020, positive): “In these works, the losses are inspired by noise contrastive estimation [13, 34] or N-pair losses [45].”
Relational Knowledge Distillation (2019, background): “Note that 2 normalization for embedding is widely used in deep metric learning [7, 13, 21, 22, 34, 41].”
Sampling Matters in Deep Embedding Learning (2017, background): “pair [28] use a more costly loss function that is defined on all pairs in a batch.”
Improving Generalization via Scalable Neighborhood Component Analysis (2018, background): “Metric learning learning encourages the minimization of intra-class variations and the maximization inter-class variations, such as contrastive loss [1,37], triplet loss [14].”
No Fuss Distance Metric Learning Using Proxies (2017, methodology): “Based on the experimental protocol detailed in [15, 14] we evaluate retrieval at k and clustering quality on data from unseen classes on 3 datasets: CUB200-2011 [17], Cars196 [6], and Stanford Online Products [8].”
Learning Metrics From Teachers: Compact Networks for Image Embedding (2019, background): “However, both hard and semi-hard negative mining have a high computational cost, which led several authors to limit the hard negative mining process to the current mini-batch [18, 31, 32, 36].”
Poisoning and Backdooring Contrastive Learning (2021, methodology): “In its most general deﬁnition, contrastive learning (Chopra et al., 2005; Hadsell et al., 2006; Sohn, 2016; Oord et al., 2018) constructs an embedding function f : X → E that maps objects of one type (e.g., images) into an embedding space so that “similar” objects have close embeddings under a…”