Towards Robust Face Recognition with Comprehensive Search

TL;DR

This paper introduces a comprehensive search framework that jointly optimizes data cleaning, architecture, and loss functions for face recognition, leading to significant performance improvements over existing methods.

Contribution

It is the first to unify the search for all three aspects, revealing their tight coupling and providing a flexible reinforcement learning-based approach for joint optimization.

Findings

Outperforms expert algorithms on large-scale benchmarks.

Optimal designs involve more challenging datasets and loss functions.

Joint search improves robustness and accuracy of face recognition.

Abstract

Data cleaning, architecture, and loss function design are important factors contributing to high-performance face recognition. Previously, the research community tries to improve the performance of each single aspect but failed to present a unified solution on the joint search of the optimal designs for all three aspects. In this paper, we for the first time identify that these aspects are tightly coupled to each other. Optimizing the design of each aspect actually greatly limits the performance and biases the algorithmic design. Specifically, we find that the optimal model architecture or loss function is closely coupled with the data cleaning. To eliminate the bias of single-aspect research and provide an overall understanding of the face recognition model design, we first carefully design the search space for each aspect, then a comprehensive search method is introduced to jointly search optimal data cleaning, architecture, and loss function design. In our framework, we make the proposed comprehensive search as flexible as possible, by using an innovative reinforcement learning based approach. Extensive experiments on million-level face recognition benchmarks demonstrate the effectiveness of our newly-designed search space for each aspect and the comprehensive search. We outperform expert algorithms developed for each single research track by large margins. More importantly, we analyze the difference between our searched optimal design and the independent design of the single factors. We point out that strong models tend to optimize with more difficult training datasets and loss functions. Our empirical study can provide guidance in future research towards more robust face recognition systems.