kNet: A Deep kNN Network To Handle Label Noise

TL;DR

kNet is a neural network designed to emulate kNN for label noise robustness, reducing memory needs and inference complexity while improving performance across various noise conditions.

Contribution

This paper introduces kNet, a neural network that learns to approximate kNN, eliminating the need for storing training data and enabling efficient inference.

Findings

kNet effectively approximates kNN with large k values.

kNet improves accuracy by up to 3% over preliminary networks.

kNet is robust across different label noise regimes.

Abstract

Deep Neural Networks require large amounts of labeled data for their training. Collecting this data at scale inevitably causes label noise.Hence,the need to develop learning algorithms that are robust to label noise. In recent years, k Nearest Neighbors (kNN) emerged as a viable solution to this problem. Despite its success, kNN is not without its problems. Mainly, it requires a huge memory footprint to store all the training samples and it needs an advanced data structure to allow for fast retrieval of the relevant examples, given a query sample. We propose a neural network, termed kNet, that learns to perform kNN. Once trained, we no longer need to store the training data, and processing a query sample is a simple matter of inference. To use kNet, we first train a preliminary network on the data set, and then train kNet on the penultimate layer of the preliminary network.We find that kNet gives a smooth approximation of kNN,and cannot handle the sharp label changes between samples that kNN can exhibit. This indicates that currently kNet is best suited to approximate kNN with a fairly large k. Experiments on two data sets show that this is the regime in which kNN works best,and can therefore be replaced by kNet.In practice, kNet consistently improve the results of all preliminary networks, in all label noise regimes, by up to 3%.