Deep Learning Forecasting

Deep learning forecasting leverages neural networks to predict future values in time series data, aiming for improved accuracy and the ability to handle complex patterns. Current research emphasizes enhancing model interpretability through techniques like counterfactual explanations, addressing challenges posed by limited data or high-dimensional datasets using methods such as matrix factorization and retrieval-based approaches, and incorporating domain-specific knowledge to improve forecasting in areas like solar energy prediction and online retail. These advancements are significant for various applications, offering more accurate predictions in diverse fields and providing valuable insights into the underlying processes driving time series behavior.

Papers

October 12, 2023

Counterfactual Explanations for Time Series Forecasting
Zhendong Wang, Ioanna Miliou, Isak Samsten, Panagiotis Papapetrou
Time Series Forecasting Counterfactual Explanation Counterfactual Generation Deep Time Series Forecasting Deep Learning Forecasting

May 23, 2023

Deep Learning based Forecasting: a case study from the online fashion industry
Manuel Kunz, Stefan Birr, Mones Raslan, Lei Ma, Zhen Li, Adele Gouttes, Mateusz Koren, Tofigh Naghibi, Johannes Stephan, Mariia Bulycheva, Matthias Grzeschik, Armin Kekić, Michael Narodovitch, Kashif Rasul, Julian Sieber, Tim Januschowski
Deep Learning Case Study Data Driven Time Series Forecasting State of the Art Forecasting Demand Forecasting Deep Learning Forecasting

September 27, 2022

Retrieval Based Time Series Forecasting
Baoyu Jing, Si Zhang, Yada Zhu, Bin Peng, Kaiyu Guan, Andrew Margenot, Hanghang Tong
Time Series Forecasting Retrieval Augmented Time Series Imputation Time sEries Deep Learning Forecasting

May 7, 2022

Deep learning for spatio-temporal forecasting -- application to solar energy
Vincent Le Guen
Deep Learning Application Proficiency Spatio Temporal Forecasting Solar Power Solar Generation Deep Learning Forecasting

December 15, 2021

Optimal Latent Space Forecasting for Large Collections of Short Time Series Using Temporal Matrix Factorization
Himanshi Charotia, Abhishek Garg, Gaurav Dhama, Naman Maheshwari
Latent Space Time Series Forecasting High Dimensional Time Series Forecasting Method Temporal Latent Large Collection Deep Learning Forecasting

Deep Learning Forecasting

Papers

Counterfactual Explanations for Time Series Forecasting

Deep Learning based Forecasting: a case study from the online fashion industry

Retrieval Based Time Series Forecasting

Deep learning for spatio-temporal forecasting -- application to solar energy

Optimal Latent Space Forecasting for Large Collections of Short Time Series Using Temporal Matrix Factorization