# Learning a Metric Embedding for Face Recognition using the Multibatch Method

arXiv: https://arxiv.org/abs/1605.07270 · HTML: https://mlsanity.com/paper/1605.07270/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 May 2016.

## Built on by (most cited first, 7 of 7)

- [Scheduled Sampling in Vision-Language Pretraining with Decoupled Encoder-Decoder Network](https://mlsanity.com/paper/2101.11562/) (2021)
- [Latent Space Explanation by Intervention](https://mlsanity.com/paper/2112.04895/) (2021)
- [RSTAM: An Effective Black-Box Impersonation Attack on Face Recognition using a Mobile and Compact Printer](https://mlsanity.com/paper/2206.12590/) (2022)
- [Recent Advances in Deep Learning Techniques for Face Recognition](https://mlsanity.com/paper/2103.10492/) (2021)
- [Improving Deep Metric Learning by Divide and Conquer](https://mlsanity.com/paper/2109.04003/) (2021)
- [A Functional Information Perspective on Model Interpretation](https://mlsanity.com/paper/2206.05700/) (2022)
- [Human Face Recognition from Part of a Facial Image based on Image Stitching](https://mlsanity.com/paper/2203.05601/) (2022)

## Abstract (arXiv)

This work is motivated by the engineering task of achieving a near state-of-the-art face recognition on a minimal computing budget running on an embedded system. Our main technical contribution centers around a novel training method, called Multibatch, for similarity learning, i.e., for the task of generating an invariant "face signature" through training pairs of "same" and "not-same" face images. The Multibatch method first generates signatures for a mini-batch of k face images and then constructs an unbiased estimate of the full gradient by relying on all k2-k pairs from the mini-batch. We prove that the variance of the Multibatch estimator is bounded by O(1/k2), under some mild conditions. In contrast, the standard gradient estimator that relies on random k/2 pairs has a variance of order 1/k. The smaller variance of the Multibatch estimator significantly speeds up the convergence rate of stochastic gradient descent. Using the Multibatch method we train a deep convolutional neural network that achieves an accuracy of 98.2% on the LFW benchmark, while its prediction runtime takes only 30msec on a single ARM Cortex A9 core. Furthermore, the entire training process took only 12 hours on a single Titan X GPU.