Robust Forecasting

Robust time series forecasting aims to develop prediction models that accurately forecast future values even when faced with noisy, incomplete, or outlier-ridden data. Current research emphasizes improving model robustness through techniques like variational mode decomposition (VMD) to handle data volatility, novel loss functions and resampling methods to mitigate the impact of outliers, and multimodal fusion approaches that integrate diverse data sources (e.g., satellite imagery, sensor data, visual information) to enhance prediction accuracy. These advancements are crucial for reliable forecasting in diverse applications, ranging from financial markets and energy management to healthcare and robotic control, where accurate predictions are essential for effective decision-making.

Papers

October 22, 2024

September 27, 2024

Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective
Chengsen Wang, Qi Qi, Jingyu Wang, Haifeng Sun, Zirui Zhuang, Jinming Wu, Jianxin Liao
Time Series Forecasting Real Power Robust Prediction Forecasting Method Global Local Timestamp Annotation Robust Forecasting

August 28, 2024

Variational Mode Decomposition and Linear Embeddings are What You Need For Time-Series Forecasting
Hafizh Raihan Kurnia Putra, Novanto Yudistira, Tirana Noor Fatyanosa
Time Series Forecasting Multivariate Forecasting Variational Mode Decomposition Robust Forecasting

June 19, 2024

Robust Time Series Forecasting with Non-Heavy-Tailed Gaussian Loss-Weighted Sampler
Jiang You, Arben Cela, René Natowicz, Jacob Ouanounou, Patrick Siarry
Multivariate Time Series Forecasting Weighted Loss Robust Forecasting Tailed Loss

June 12, 2024

Rating Multi-Modal Time-Series Forecasting Models (MM-TSFM) for Robustness Through a Causal Lens
Kausik Lakkaraju, Rachneet Kaur, Zhen Zeng, Parisa Zehtabi, Sunandita Patra, Biplav Srivastava, Marco Valtorta
Native Robustness Time Series Forecasting Human Rating Robust Forecasting Forecasting Analysis

May 30, 2024

Urban Air Pollution Forecasting: a Machine Learning Approach leveraging Satellite Observations and Meteorological Forecasts
Giacomo Blanco, Luca Barco, Lorenzo Innocenti, Claudio Rossi
Machine Learning Model Machine Learning Approach Weather Forecasting Air Pollution Satellite Observation Air Quality Forecasting Pollution Level Robust Forecasting

February 3, 2024

RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies
Hao Cheng, Qingsong Wen, Yang Liu, Liang Sun
Product Design Time Series Forecasting Theoretical Understanding Unlabeled Anomaly Robust Loss Time Series Forecasting Method Robust Forecasting

September 7, 2023

A Natural Gas Consumption Forecasting System for Continual Learning Scenarios based on Hoeffding Trees with Change Point Detection Mechanism
Radek Svoboda, Sebastian Basterrech, Jędrzej Kozal, Jan Platoš, Michał Woźniak
Change Point Detection Forecasting Model Continual Learning Scenario Robust Forecasting Forecast Error Hoeffding Tree

March 27, 2023

CoRe-Sleep: A Multimodal Fusion Framework for Time Series Robust to Imperfect Modalities
Konstantinos Kontras, Christos Chatzichristos, Huy Phan, Johan Suykens, Maarten De Vos
Multimodal Fusion Different Modality Sleep Staging Sleep Stage Classification Robust Forecasting Multimodal Fusion Network

September 22, 2022

Robust Forecasting for Robotic Control: A Game-Theoretic Approach
Shubhankar Agarwal, David Fridovich-Keil, Sandeep P. Chinchali
Game Theoretic Robot Control Two Player Zero Sum Robust Forecasting

April 18, 2022

Multi-scale Anomaly Detection for Big Time Series of Industrial Sensors
Quan Ding, Shenghua Liu, Bin Zhou, Huawei Shen, Xueqi Cheng
Time Series Adversarial Regularization Industrial Monitoring Reconstruction Based Anomaly Detection Robust Forecasting

January 31, 2022

Continuous Forecasting via Neural Eigen Decomposition
Stav Belogolovsky, Ido Greenberg, Danny Eitan, Shie Mannor
Differential Equation Piecewise Linear Neural SDEs Robust Forecasting Neural Eigenfunctions

Robust Forecasting

Papers

LiNo: Advancing Recursive Residual Decomposition of Linear and Nonlinear Patterns for Robust Time Series Forecasting

xLSTM-Mixer: Multivariate Time Series Forecasting by Mixing via Scalar Memories

Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective

Variational Mode Decomposition and Linear Embeddings are What You Need For Time-Series Forecasting

Robust Time Series Forecasting with Non-Heavy-Tailed Gaussian Loss-Weighted Sampler

Rating Multi-Modal Time-Series Forecasting Models (MM-TSFM) for Robustness Through a Causal Lens

Urban Air Pollution Forecasting: a Machine Learning Approach leveraging Satellite Observations and Meteorological Forecasts

RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies

A Natural Gas Consumption Forecasting System for Continual Learning Scenarios based on Hoeffding Trees with Change Point Detection Mechanism

CoRe-Sleep: A Multimodal Fusion Framework for Time Series Robust to Imperfect Modalities

Robust Forecasting for Robotic Control: A Game-Theoretic Approach

Multi-scale Anomaly Detection for Big Time Series of Industrial Sensors

Continuous Forecasting via Neural Eigen Decomposition