Reproducing BowNet: Learning Representations by Predicting Bags of Visual Words

TL;DR

This paper attempts to reproduce the results of BowNet, a self-supervised learning method using bag-of-visual-words descriptors, but faces challenges in achieving the original reported accuracy improvements.

Contribution

It provides an implementation effort to reproduce BowNet's results and discusses potential reasons for the difficulties encountered.

Findings

Failed to replicate the original accuracy improvements

Identified potential factors affecting reproducibility

Highlights challenges in reproducing SSL methods

Abstract

This work aims to reproduce results from the CVPR 2020 paper by Gidaris et al. Self-supervised learning (SSL) is used to learn feature representations of an image using an unlabeled dataset. This work proposes to use bag-of-words (BoW) deep feature descriptors as a self-supervised learning target to learn robust, deep representations. BowNet is trained to reconstruct the histogram of visual words (ie. the deep BoW descriptor) of a reference image when presented a perturbed version of the image as input. Thus, this method aims to learn perturbation-invariant and context-aware image features that can be useful for few-shot tasks or supervised downstream tasks. In the paper, the author describes BowNet as a network consisting of a convolutional feature extractor $\Phi(\cdot)$ and a Dense-softmax layer $\Omega(\cdot)$ trained to predict BoW features from images. After BoW training, the features of $\Phi$ are used in downstream tasks. For this challenge we were trying to build and train a network that could reproduce the CIFAR-100 accuracy improvements reported in the original paper. However, we were unsuccessful in reproducing an accuracy improvement comparable to what the authors mentioned. This could be for a variety of factors and we believe that time constraints were the primary bottleneck.