Revisiting Training Strategies and Generalization Performance in Deep Metric Learning

Abstract

Deep Metric Learning (DML) is arguably one of the most influential lines of research for learning visual similarities with many proposed approaches every year. Although the field benefits from the rapid progress, the divergence in training protocols, architectures, and parameter choices make an unbiased comparison difficult. To provide a consistent reference point, we revisit the most widely used DML objective functions and conduct a study of the crucial parameter choices as well as the commonly neglected mini-batch sampling process. Under consistent comparison, DML objectives show much higher saturation than indicated by literature. Further based on our analysis, we uncover a correlation between the embedding space density and compression to the generalization performance of DML models. Exploiting these insights, we propose a simple, yet effective, training regularization to reliably boost the performance of ranking-based DML models on various standard benchmark datasets. Code and a publicly accessible WandB-repo are available at https://github.com/Confusezius/Revisiting_Deep_Metric_Learning_PyTorch.

ICML 2020. Main paper 8.25 pages, 26 pages total

References in corpus (18)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• Opening the Black Box of Deep Neural Networks via Information
• Don't Decay the Learning Rate, Increase the Batch Size
• On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
• Manifold Mixup: Better Representations by Interpolating Hidden States
• MINE: Mutual Information Neural Estimation
• Deep Variational Information Bottleneck
• Learning Deep Embeddings with Histogram Loss
• Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observations
• SoftTriple Loss: Deep Metric Learning Without Triplet Sampling
• Robustness and Generalization for Metric Learning
• InfoBot: Transfer and Exploration via the Information Bottleneck
• Small-GAN: Speeding Up GAN Training Using Core-sets
• Sharing Matters for Generalization in Deep Metric Learning
• Supervised Metric Learning with Generalization Guarantees

Cited by in corpus (11)

• COVID-19 Image Data Collection: Prospective Predictions Are the Future
• Sharing Matters for Generalization in Deep Metric Learning
• Multi-task Batch Reinforcement Learning with Metric Learning
• Adversarial Attack and Defense in Deep Ranking
• S2SD: Simultaneous Similarity-based Self-Distillation for Deep Metric Learning
• Few-shot Intent Classification and Slot Filling with Retrieved Examples
• Uniform Priors for Data-Efficient Transfer
• Proxy Synthesis: Learning with Synthetic Classes for Deep Metric Learning
• Color Variants Identification in Fashion e-commerce via Contrastive Self-Supervised Representation Learning
• Learning Similarity Metrics for Volumetric Simulations with Multiscale CNNs
• Adaptive additive classification-based loss for deep metric learning