XTSFormer: Cross-Temporal-Scale Transformer for Irregular Time Event Prediction
Primary Area: general machine learning (i.e., none of the above)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: transformer, event sequence prediction, irregular time
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: Our paper introduces the XTSFormer, a novel model optimized for predicting irregularly timed events, which can uniquely captures the multi-scale interactions and cyclical nature of time.
Abstract: Event prediction aims to forecast the time and type of a future event based on a historical event sequence. Despite its significance, several challenges exist, including the irregularity of time intervals between events, cycles, periodicity, and the complex multi-scale nature of event interactions, as well as the potentially high computational costs for long event sequences. However, current neural temporal point processes (TPPs) methods do not capture the multi-scale nature of event interactions, which is common in many real-world applications such as clinical event data. To address these issues, we propose the cross-temporal-scale transformer (XTSFormer), designed specifically for irregularly timed event data. Our model comprises two vital components: a novel Feature-based Cycle-aware Positional Encoding (FCPE) that adeptly captures the cyclical nature of time, and a hierarchical multi-scale temporal attention mechanism. These scales are determined by a bottom-up clustering algorithm. Extensive experiments on several real-world datasets show that our XTSFormer outperforms several baseline methods in prediction performance.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 6853
Loading