Curriculum Learning for Multi-Task Classification of Visual Attributes

TL;DR

This paper presents a novel curriculum learning approach for multi-task visual attribute classification, grouping tasks by correlation to improve learning efficiency and accuracy, achieving state-of-the-art results.

Contribution

It introduces a method that combines multi-task and curriculum learning by grouping tasks based on correlation and transferring knowledge between groups.

Findings

Faster convergence compared to standard multi-task learning.

Achieves state-of-the-art results on multiple datasets.

Effective grouping based on task correlation improves performance.

Abstract

Visual attributes, from simple objects (e.g., backpacks, hats) to soft-biometrics (e.g., gender, height, clothing) have proven to be a powerful representational approach for many applications such as image description and human identification. In this paper, we introduce a novel method to combine the advantages of both multi-task and curriculum learning in a visual attribute classification framework. Individual tasks are grouped based on their correlation so that two groups of strongly and weakly correlated tasks are formed. The two groups of tasks are learned in a curriculum learning setup by transferring the acquired knowledge from the strongly to the weakly correlated. The learning process within each group though, is performed in a multi-task classification setup. The proposed method learns better and converges faster than learning all the tasks in a typical multi-task learning paradigm. We demonstrate the effectiveness of our approach on the publicly available, SoBiR, VIPeR and PETA datasets and report state-of-the-art results across the board.