Input Time Series

Input time series analysis focuses on extracting meaningful information and making predictions from sequential data across various domains. Current research emphasizes developing sophisticated neural network architectures, including Transformers, MLP-Mixers, and hypercomplex networks, as well as leveraging techniques like self-supervised learning and co-training to improve accuracy and efficiency, particularly for long and multivariate time series. These advancements are driving improvements in forecasting accuracy across diverse applications such as finance, healthcare, and environmental monitoring, while also addressing challenges like noise and data sparsity. The field is also seeing exploration of novel approaches such as reprogramming large language models for time series tasks and the development of more efficient and interpretable models.

Papers

May 22, 2024

Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers
Xin Cheng, Xiuying Chen, Shuqi Li, Di Luo, Xun Wang, Dongyan Zhao, Rui Yan
2 Dimensional Time Series Prediction Long Term Time Series Forecasting Vanilla Transformer Grid Based Input Time Series Covariate Representation

January 9, 2024

Hypercomplex neural network in time series forecasting of stock data
Radosław Kycia, Agnieszka Niemczynowicz
Neural Network Long Short Term Memory Time Series Forecasting Time Series Prediction Indian Stock Market Hypercomplex Algebra Input Time Series

January 3, 2024

Kernel-U-Net: Multivariate Time Series Forecasting using Custom Kernels
Jiang You, Arben Cela, René Natowicz, Jacob Ouanounou, Patrick Siarry
Time Series Forecasting U Net Multivariate Time Series Forecasting Input Time Series

November 19, 2023

Self-Distilled Representation Learning for Time Series
Felix Pieper, Konstantin Ditschuneit, Martin Genzel, Alexandra Lindt, Johannes Otterbach
Contrastive Learning Self Supervised Time Series Self Supervised Learning Method Input Time Series

October 18, 2023

A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis
Shuhan Zhong, Sizhe Song, Weipeng Zhuo, Guanyao Li, Yang Liu, S. -H. Gary Chan
Multi Scale Time Series Analysis MLP Mixer Input Time Series Inter Series

October 3, 2023

Time-LLM: Time Series Forecasting by Reprogramming Large Language Models
Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, Xiaoming Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen
Large Language Model Time Series Forecasting Input Time Series

August 24, 2023

A Co-training Approach for Noisy Time Series Learning
Weiqi Zhang, Jianfeng Zhang, Jia Li, Fugee Tsung
Contrastive Learning Time Series Time Series Representation Learning Real World Time Series Input Time Series Time Series Benchmark

June 14, 2023

GCformer: An Efficient Framework for Accurate and Scalable Long-Term Multivariate Time Series Forecasting
YanJun Zhao, Ziqing Ma, Tian Zhou, Liang Sun, Mengni Ye, Yi Qian
Transformer Based Model Multivariate Time Series Forecasting Efficient Framework Input Time Series Convolutional Branch 1D Kernel

May 8, 2023

Mlinear: Rethink the Linear Model for Time-series Forecasting
Wei Li, Xiangxu Meng, Chuhao Chen, Jianing Chen
Time Series Forecasting Linear Model Time Series Datasets Channel Prediction Input Time Series

March 22, 2023

TSI-GAN: Unsupervised Time Series Anomaly Detection using Convolutional Cycle-Consistent Generative Adversarial Networks
Shyam Sundar Saravanan, Tie Luo, Mao Van Ngo
Anomaly Detection Unsupervised Anomaly Detection C Gan Temporal Pattern Unsupervised Time Series Anomaly Detection Input Time Series Cycle Consistent Generative Adversarial Network

January 21, 2023

Ti-MAE: Self-Supervised Masked Time Series Autoencoders
Zhe Li, Zhongwen Rao, Lujia Pan, Pengyun Wang, Zenglin Xu
Time Series Forecasting Multivariate Time Series Forecasting Masked AutoEncoder Sce Mae Transformer Based Generative Input Time Series

June 18, 2022

Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting
Zezhi Shao, Zhao Zhang, Fei Wang, Yongjun Xu
Time Series Multivariate Time Series Multivariate Time Series Forecasting Temporal Graph Neural Network Input Time Series Multivariate Time

April 26, 2022

Forecasting foreign exchange rates with regression networks tuned by Bayesian optimization
Linwei Li, Paul-Amaury Matt, Christian Heumann
Bayesian Optimization Multi Step Input Time Series Ornstein Uhlenbeck Process Regression Network Varying Coefficient

April 19, 2022

EXIT: Extrapolation and Interpolation-based Neural Controlled Differential Equations for Time-series Classification and Forecasting
Sheo Yon Jhin, Jaehoon Lee, Minju Jo, Seungji Kook, Jinsung Jeon, Jihyeon Hyeong, Jayoung Kim, Noseong Park
Time Series Classification Differential Equation State of the Art Forecasting Traditional RNNs Extrapolation Framework Sampled Time Series Input Time Series Interpolation Learning Extrinsic Manipulation

March 25, 2022

HYDRA: Competing convolutional kernels for fast and accurate time series classification
Angus Dempster, Daniel F. Schmidt, Geoffrey I. Webb
Time Series Time Series Classification Convolutional Kernel Hydra MDP Input Time Series Dictionary Based

February 21, 2022

Recurrent Auto-Encoder With Multi-Resolution Ensemble and Predictive Coding for Multivariate Time-Series Anomaly Detection
Heejeong Choi, Subin Kim, Pilsung Kang
Time Series Data Predictive Coding Auto Encoder Multivariate Time Series Anomaly Detection Input Time Series

Input Time Series

Papers

Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers

Hypercomplex neural network in time series forecasting of stock data

Kernel-U-Net: Multivariate Time Series Forecasting using Custom Kernels

Self-Distilled Representation Learning for Time Series

A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis

Time-LLM: Time Series Forecasting by Reprogramming Large Language Models

A Co-training Approach for Noisy Time Series Learning

GCformer: An Efficient Framework for Accurate and Scalable Long-Term Multivariate Time Series Forecasting

Mlinear: Rethink the Linear Model for Time-series Forecasting

TSI-GAN: Unsupervised Time Series Anomaly Detection using Convolutional Cycle-Consistent Generative Adversarial Networks

Ti-MAE: Self-Supervised Masked Time Series Autoencoders

Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting

Forecasting foreign exchange rates with regression networks tuned by Bayesian optimization

EXIT: Extrapolation and Interpolation-based Neural Controlled Differential Equations for Time-series Classification and Forecasting

HYDRA: Competing convolutional kernels for fast and accurate time series classification

Recurrent Auto-Encoder With Multi-Resolution Ensemble and Predictive Coding for Multivariate Time-Series Anomaly Detection