Feb 15, 2022 · In this paper, we systematically review Transformer schemes for time series modeling by highlighting their strengths as well as limitations.

Feb 4, 2024 · The outcome of this study is a Time Series Transformer (Timer), which is generative pre-trained by next token prediction and adapted to various downstream ...

In this paper, we propose a new time series forecasting model -- Adversarial Sparse Transformer (AST), based on Generated Adversarial Networks (GANs).

We propose LogSparse Transformer with only O(L(log L)^2) memory cost, improving forecasting accuracy for time series with fine granularity and strong long-term ...

This is an official implementation of the NeurIPS2023 paper: "Time Series as Images: Vision Transformer for Irregularly Sampled Time Series".

Mar 7, 2023 · We investigate the Transformer model to address an important sequence learning problem in finance: time series forecasting.

Aug 14, 2021 · In this work we propose for the first time a transformer-based framework for unsupervised representation learning of multivariate time series.

Jan 3, 2023 · It seems to cast doubt on the recent trend of using transformers for time series forecasting, suggesting a simple model can out perform complex transformers.

In this paper, we propose the Temporal Fusion Transformer (TFT) – an attention-based DNN architecture for multi-horizon forecasting that achieves high ...

Dec 19, 2023 · In this article, we aim to provide a brief overview of some interesting work that adapts the Transformer architecture to facilitate time series data.