Learning from Complexity: Exploring Dynamic Sample Pruning of Spatio-Temporal Training

TL;DR

This paper introduces ST-Prune, a dynamic sample pruning method that enhances training efficiency in spatio-temporal forecasting by focusing on informative samples, leading to faster convergence and improved performance.

Contribution

The paper presents a novel dynamic sample pruning technique for spatio-temporal forecasting that reduces training time without sacrificing accuracy.

Findings

Significantly accelerates training speed

Maintains or improves model performance

Demonstrates scalability and universality

Abstract

Spatio-temporal forecasting is fundamental to intelligent systems in transportation, climate science, and urban planning. However, training deep learning models on the massive, often redundant, datasets from these domains presents a significant computational bottleneck. Existing solutions typically focus on optimizing model architectures or optimizers, while overlooking the inherent inefficiency of the training data itself. This conventional approach of iterating over the entire static dataset each epoch wastes considerable resources on easy-to-learn or repetitive samples. In this paper, we explore a novel training-efficiency techniques, namely learning from complexity with dynamic sample pruning, ST-Prune, for spatio-temporal forecasting. Through dynamic sample pruning, we aim to intelligently identify the most informative samples based on the model's real-time learning state, thereby accelerating convergence and improving training efficiency. Extensive experiments conducted on real-world spatio-temporal datasets show that ST-Prune significantly accelerates the training speed while maintaining or even improving the model performance, and it also has scalability and universality.