Entropy-Driven Curriculum for Multi-Task Training in Human Mobility Prediction
This work addresses training inefficiencies and accuracy limitations in human mobility prediction for applications using big mobility data, representing an incremental improvement through integration of existing techniques.
The paper tackles the problem of inefficient training and suboptimal accuracy in human mobility prediction by introducing a unified framework combining entropy-driven curriculum learning and multi-task training, achieving state-of-the-art performance with a 2.92-fold convergence speed improvement and metrics like GEO-BLEU of 0.354.
The increasing availability of big mobility data from ubiquitous portable devices enables human mobility prediction through deep learning approaches. However, the diverse complexity of human mobility data impedes model training, leading to inefficient gradient updates and potential underfitting. Meanwhile, exclusively predicting next locations neglects implicit determinants, including distances and directions, thereby yielding suboptimal prediction results. This paper presents a unified training framework that integrates entropy-driven curriculum and multi-task learning to address these challenges. The proposed entropy-driven curriculum learning strategy quantifies trajectory predictability based on Lempel-Ziv compression and organizes training from simple to complex for faster convergence and enhanced performance. The multi-task training simultaneously optimizes the primary location prediction alongside auxiliary estimation of movement distance and direction for learning realistic mobility patterns, and improve prediction accuracy through complementary supervision signals. Extensive experiments conducted in accordance with the HuMob Challenge demonstrate that our approach achieves state-of-the-art performance on GEO-BLEU (0.354) and DTW (26.15) metrics with up to 2.92-fold convergence speed compared to training without curriculum learning.