Transfer Learning Based on AdaBoost for Feature Selection from Multiple ConvNet Layer Features

TL;DR

This paper demonstrates that combining multiple ConvNet layer features with AdaBoost-based feature selection improves transfer learning performance, especially when source and target tasks are more dissimilar.

Contribution

It introduces a novel approach using AdaBoost with single stumps to select relevant features from concatenated ConvNet layers for transfer learning.

Findings

Multiple ConvNet layer features outperform single layer features.

AdaBoost effectively selects useful features from complex feature spaces.

Performance gains increase with greater source-target task dissimilarity.

Abstract

Convolutional Networks (ConvNets) are powerful models that learn hierarchies of visual features, which could also be used to obtain image representations for transfer learning. The basic pipeline for transfer learning is to first train a ConvNet on a large dataset (source task) and then use feed-forward units activation of the trained ConvNet as image representation for smaller datasets (target task). Our key contribution is to demonstrate superior performance of multiple ConvNet layer features over single ConvNet layer features. Combining multiple ConvNet layer features will result in more complex feature space with some features being repetitive. This requires some form of feature selection. We use AdaBoost with single stumps to implicitly select only distinct features that are useful towards classification from concatenated ConvNet features. Experimental results show that using multiple ConvNet layer activation features instead of single ConvNet layer features consistently will produce superior performance. Improvements becomes significant as we increase the distance between source task and the target task.