LGAIOct 31, 2024

LATST: Are Transformers Necessarily Complex for Time-Series Forecasting

arXiv:2410.23749v91 citationsh-index: 1
Originality Incremental advance
AI Analysis

This work addresses a key bottleneck in time-series forecasting for researchers and practitioners, offering an incremental improvement over existing Transformer models.

The paper tackles the underperformance of Transformer-based models in multivariate long-term time-series forecasting compared to simpler linear baselines by introducing LATST, a novel approach that mitigates entropy collapse and training instability, achieving competitive performance with fewer parameters on some datasets.

Transformer-based architectures have achieved remarkable success in natural language processing and computer vision. However, their performance in multivariate long-term forecasting often falls short compared to simpler linear baselines. Previous research has identified the traditional attention mechanism as a key factor limiting their effectiveness in this domain. To bridge this gap, we introduce LATST, a novel approach designed to mitigate entropy collapse and training instability common challenges in Transformer-based time series forecasting. We rigorously evaluate LATST across multiple real-world multivariate time series datasets, demonstrating its ability to outperform existing state-of-the-art Transformer models. Notably, LATST manages to achieve competitive performance with fewer parameters than some linear models on certain datasets, highlighting its efficiency and effectiveness.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes