Deep Active Learning Using Barlow Twins

TL;DR

This paper introduces Deep Active Learning using Barlow Twins (DALBT), a novel approach combining self-supervised learning with active learning to efficiently select informative samples for training CNNs, reducing annotation costs.

Contribution

It proposes a new active learning method that leverages Barlow Twins' self-supervised framework to improve sample selection for CNN training.

Findings

DALBT achieves high accuracy with fewer labeled samples.

The method effectively encodes invariance to distortions.

It outperforms traditional active learning approaches.

Abstract

The generalisation performance of a convolutional neural networks (CNN) is majorly predisposed by the quantity, quality, and diversity of the training images. All the training data needs to be annotated in-hand before, in many real-world applications data is easy to acquire but expensive and time-consuming to label. The goal of the Active learning for the task is to draw most informative samples from the unlabeled pool which can used for training after annotation. With total different objective, self-supervised learning which have been gaining meteoric popularity by closing the gap in performance with supervised methods on large computer vision benchmarks. self-supervised learning (SSL) these days have shown to produce low-level representations that are invariant to distortions of the input sample and can encode invariance to artificially created distortions, e.g. rotation, solarization, cropping etc. self-supervised learning (SSL) approaches rely on simpler and more scalable frameworks for learning. In this paper, we unify these two families of approaches from the angle of active learning using self-supervised learning mainfold and propose Deep Active Learning using BarlowTwins(DALBT), an active learning method for all the datasets using combination of classifier trained along with self-supervised loss framework of Barlow Twins to a setting where the model can encode the invariance of artificially created distortions, e.g. rotation, solarization, cropping etc.