Corgi^2: A Hybrid Offline-Online Approach To Storage-Aware Data Shuffling For SGD

TL;DR

This paper introduces a hybrid offline-online data shuffling method for SGD that balances efficiency and randomness, improving training on large cloud-stored datasets.

Contribution

It proposes a novel two-step shuffling strategy combining offline and online methods, enhancing data access efficiency while maintaining convergence performance.

Findings

Achieves similar convergence to fully random shuffling.

Reduces data access costs for large datasets.

Demonstrates practical benefits through experiments.

Abstract

When using Stochastic Gradient Descent (SGD) for training machine learning models, it is often crucial to provide the model with examples sampled at random from the dataset. However, for large datasets stored in the cloud, random access to individual examples is often costly and inefficient. A recent work \cite{corgi}, proposed an online shuffling algorithm called CorgiPile, which greatly improves efficiency of data access, at the cost some performance loss, which is particularly apparent for large datasets stored in homogeneous shards (e.g., video datasets). In this paper, we introduce a novel two-step partial data shuffling strategy for SGD which combines an offline iteration of the CorgiPile method with a subsequent online iteration. Our approach enjoys the best of both worlds: it performs similarly to SGD with random access (even for homogenous data) without compromising the data access efficiency of CorgiPile. We provide a comprehensive theoretical analysis of the convergence properties of our method and demonstrate its practical advantages through experimental results.