Grouped self-attention mechanism for a memory-efficient Transformer

Bumjun Jung, YUSUKE Mukuta, Tatsuya Harada

Keywords: time-series data, Transformer, memory-efficient structure, self-attention, computational and space complexity, long-range dependency

Abstract: Time-series data analysis is important because numerous real-world tasks such as forecasting weather, electricity consumption, and stock market involve predicting data that vary over time. Time-series data are generally recorded over a long period of observation with long sequences owing to their periodic characteristics and long-range dependencies over time. Thus, capturing long-range dependency is an important factor in time-series data forecasting. To solve these problems, we proposed two novel modules, Grouped Self-Attention (GSA) and Compressed Cross-Attention (CCA). With both modules, we achieved a computational space and time complexity of order $O(l)$ with a sequence length $l$ under small hyperparameter limitations, and can capture locality while considering global information. The results of experiments conducted on time-series datasets show that our proposed model efficiently exhibited reduced computational complexity and performance comparable to or better than existing methods.

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.

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

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning

TL;DR: We proposed Grouped Self-Attention (GSA) and Compressed Cross-Attention (CCA) modules to solve the quadratic order problem on self-attention mechanism of Transformer.