Curriculum generation using Autoencoder based continuous optimization

TL;DR

This paper introduces a novel curriculum generation method called Training Sequence Optimization (TSO) that uses autoencoder-based continuous optimization to improve training data sequencing, leading to better model performance.

Contribution

The paper presents a simple, efficient continuous optimization approach for curriculum learning using auto-encoding, outperforming existing reinforcement learning methods.

Findings

Achieved a 2% accuracy gain over random strategies on CIFAR datasets.

Outperformed state-of-the-art curriculum learning algorithms.

Demonstrated effectiveness with gradient-based optimization in training sequence design.

Abstract

Research in Curriculum Learning has shown better performance on the task by optimizing the sequence of the training data. Recent works have focused on using complex reinforcement learning techniques to find the optimal data ordering strategy to maximize learning for a given network. In this paper, we present a simple yet efficient technique based on continuous optimization trained with auto-encoding procedure. We call this new approach Training Sequence Optimization (TSO). With a usual encoder-decoder setup we try to learn the latent space continuous representation of the training strategy and a predictor network is used on the continuous representation to predict the accuracy of the strategy on the fixed network architecture. The performance predictor and encoder enable us to perform gradient-based optimization by gradually moving towards the latent space representation of training data ordering with potentially better accuracy. We show an empirical gain of 2AP with our generated optimal curriculum strategy over the random strategy using the CIFAR-100 and CIFAR-10 datasets and have better boosts than the existing state-of-the-art CL algorithms.