Decoupling "when to update" from "how to update"

TL;DR

This paper introduces a meta algorithm that separates the timing of updates from the update method to effectively handle noisy labels in data mining for gender classification, achieving state-of-the-art results.

Contribution

The paper proposes a novel meta algorithm that decouples 'when to update' from 'how to update' to improve learning from noisy, multi-source data.

Findings

Achieves state-of-the-art results in gender classification.

Effectively handles noisy labels from diverse data sources.

Provides analysis of convergence properties.

Abstract

Deep learning requires data. A useful approach to obtain data is to be creative and mine data from various sources, that were created for different purposes. Unfortunately, this approach often leads to noisy labels. In this paper, we propose a meta algorithm for tackling the noisy labels problem. The key idea is to decouple "when to update" from "how to update". We demonstrate the effectiveness of our algorithm by mining data for gender classification by combining the Labeled Faces in the Wild (LFW) face recognition dataset with a textual genderizing service, which leads to a noisy dataset. While our approach is very simple to implement, it leads to state-of-the-art results. We analyze some convergence properties of the proposed algorithm.