Efficient Time Series Forecasting

Efficient time series forecasting aims to accurately predict future values in sequential data while minimizing computational cost, a crucial need across diverse fields. Current research emphasizes developing faster algorithms, including non-autoregressive models and automated machine learning frameworks that optimize model selection and pipeline construction, as well as leveraging pre-trained large language models and adapting them for time series data. These advancements improve forecasting accuracy and efficiency for applications ranging from energy demand prediction to weather forecasting and network optimization, impacting both scientific understanding and practical decision-making.

Papers

November 7, 2024

EffiCANet: Efficient Time Series Forecasting with Convolutional Attention
Xinxing Zhou, Jiaqi Ye, Shubao Zhao, Ming Jin, Chengyi Yang, Yanlong Wen, Xiaojie Yuan
Deep Convolutional Neural Network Human Attention Group Convolution Large Kernel Convolution Superior Forecasting Efficient Time Series Forecasting

October 15, 2024

UmambaTSF: A U-shaped Multi-Scale Long-Term Time Series Forecasting Method Using Mamba
Li Wu, Wenbin Pei, Jiulong Jiao, Qiang Zhang
Practical Method Mamba in Mamba Multivariate Time Series Forecasting Long Term Time Series Forecasting Sequence Representation Efficient Time Series Forecasting

June 7, 2024

TDT Loss Takes It All: Integrating Temporal Dependencies among Targets into Non-Autoregressive Time Series Forecasting
Qi Xiong, Kai Tang, Minbo Ma, Jie Xu, Tianrui Li
Time Series Forecasting Temporal Dependency Non Autoregressive Multiple Target Efficient Time Series Forecasting

December 13, 2023

auto-sktime: Automated Time Series Forecasting
Marc-André Zöller, Marius Lindauer, Marco F. Huber
Time Series Forecasting Real World Time Series Efficient Time Series Forecasting Forecasting Pipeline

August 16, 2023

LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters
Ching Chang, Wei-Yao Wang, Wen-Chih Peng, Tien-Fu Chen
Large Language Model Fine Tuning Multi Scale Time Series Forecasting Pre Trained Large Language Model Downstream Forecasting Task Efficient Time Series Forecasting

June 9, 2023

QBSD: Quartile-Based Seasonality Decomposition for Cost-Effective RAN KPI Forecasting
Ebenezer RHP Isaac, Bulbul Singh
State of the Art Forecasting Forecasting Performance Real Time Prediction Forecasting Competition Efficient Time Series Forecasting Seasonal Trend Decomposition

May 28, 2023

Adaptive Sparsity Level during Training for Efficient Time Series Forecasting with Transformers
Zahra Atashgahi, Mykola Pechenizkiy, Raymond Veldhuis, Decebal Constantin Mocanu
Training Data Transformer Megatron Decepticons Sparse Neural Network Sparse Training Forecasting Analysis Efficient Time Series Forecasting Sparsity Tradeoff Adaptive Sparsity

May 11, 2023

A Generic Approach to Integrating Time into Spatial-Temporal Forecasting via Conditional Neural Fields
Minh-Thanh Bui, Duc-Thinh Ngo, Demin Lu, Zonghua Zhang
Forecasting Model Spatio Temporal Forecasting Autoregressive Neural Network Efficient Time Series Forecasting Forecasting Algorithm Conditional Neural Field Driving Network Long Time Integration

May 27, 2022

Efficient Forecasting of Large Scale Hierarchical Time Series via Multilevel Clustering
Xing Han, Tongzheng Ren, Jing Hu, Joydeep Ghosh, Nhat Ho
Time Series Data Hierarchical Time Series Efficient Time Series Forecasting Global Clustering Level Clustering

May 10, 2022

Quality versus speed in energy demand prediction for district heating systems
Witold Andrzejewski, Jedrzej Potoniec, Maciej Drozdowski, Jerzy Stefanowski, Robert Wrembel, Paweł Stapf
Quality Issue Speed Effect Predictive Algorithm Efficient Time Series Forecasting District Heating Grid

April 1, 2022

A Global Modeling Approach for Load Forecasting in Distribution Networks
Miha Grabner, Yi Wang, Qingsong Wen, Boštjan Blažič, Vitomir Štruc
Load Forecasting Global Model Forecasting Model Smart Meter Distribution System Efficient Time Series Forecasting

November 14, 2021

Skillful Twelve Hour Precipitation Forecasts using Large Context Neural Networks
Lasse Espeholt, Shreya Agrawal, Casper Sønderby, Manoj Kumar, Jonathan Heek, Carla Bromberg, Cenk Gazen, Jason Hickey, Aaron Bell, Nal Kalchbrenner
Weather Forecasting Weather Prediction Model Precipitation Prediction Efficient Time Series Forecasting Physic Based Numerical Weather Prediction Extreme Precipitation

Efficient Time Series Forecasting

Papers

EffiCANet: Efficient Time Series Forecasting with Convolutional Attention

UmambaTSF: A U-shaped Multi-Scale Long-Term Time Series Forecasting Method Using Mamba

TDT Loss Takes It All: Integrating Temporal Dependencies among Targets into Non-Autoregressive Time Series Forecasting

auto-sktime: Automated Time Series Forecasting

LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters

QBSD: Quartile-Based Seasonality Decomposition for Cost-Effective RAN KPI Forecasting

Adaptive Sparsity Level during Training for Efficient Time Series Forecasting with Transformers

A Generic Approach to Integrating Time into Spatial-Temporal Forecasting via Conditional Neural Fields

Efficient Forecasting of Large Scale Hierarchical Time Series via Multilevel Clustering

Quality versus speed in energy demand prediction for district heating systems

A Global Modeling Approach for Load Forecasting in Distribution Networks

Skillful Twelve Hour Precipitation Forecasts using Large Context Neural Networks