AA-forecast: anomaly-aware forecast for extreme events (2022-08-21T00:00:00.000000Z)

TL;DR

This work proposes an anomaly-aware forecast framework that leverages the effects of anomalies to improve its prediction accuracy during the presence of extreme events and employs a dynamic uncertainty optimization algorithm that reduces the uncertainty of forecasts in an online manner.

Abstract

Time series models often are impacted by extreme events and anomalies, both prevalent in real-world datasets. Such models require careful probabilistic forecasts, which is vital in risk management for extreme events such as hurricanes and pandemics. However, it’s challenging to automatically detect and learn from extreme events and anomalies for large-scale datasets which often results in extra manual efforts. Here, we propose an anomaly-aware forecast framework that leverages the effects of anomalies to improve its prediction accuracy during the presence of extreme events. Our model has trained to extract anomalies automatically and incorporates them through an attention mechanism to increase the accuracy of forecasts during extreme events. Moreover, the framework employs a dynamic uncertainty optimization algorithm that reduces the uncertainty of forecasts in an online manner. The proposed framework demonstrated consistent superior accuracy with less uncertainty on three datasets with different varieties of anomalies over the current prediction models.

Authors

Haoyi Xiong

5 papers

Jiang Bian

3 papers

Jun Wang

2 papers

TL;DR

Abstract

Authors

References33 items

COVID-19 modelling by time-varying transmission rate associated with mobility trend of driving via Apple Maps

Minimizing production loss consequent to disasters using a subsidy optimization model: a pandemic case

MLP-Mixer: An all-MLP Architecture for Vision

Novelty Detection in Sequential Data by Informed Clustering and Modeling

COVID-19 and sectoral employment trends: assessing resilience in the US leisure and hospitality industry

Machine-learning-assisted fabrication: Bayesian optimization of laser-induced graphene patterning using in-situ Raman analysis

Big Bird: Transformers for Longer Sequences

Traffic transformer: Capturing the continuity and periodicity of time series for traffic forecasting

Extreme events defined—A conceptual discussion applying a complex systems approach

Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting

Unsupervised Pre-training of a Deep LSTM-based Stacked Autoencoder for Multivariate Time Series Forecasting Problems

Learning Interpretable Deep State Space Model for Probabilistic Time Series Forecasting

Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting

Tackling Climate Change with Machine Learning

Survey of Dropout Methods for Deep Neural Networks

RobustSTL: A Robust Seasonal-Trend Decomposition Algorithm for Long Time Series

Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: a time-series analysis.

Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context

Deep and Confident Prediction for Time Series at Uber

Short-term load forecasting using a two-stage sarimax model

DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks

Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

Risk-Based Assessment of Unplanned Outage Events and Costs for Combined-Cycle Plants

Risk-Based Assessment of Unplanned Outage Events and Costs for Combined-Cycle-Plants

Long Short-Term Memory

Robust Locally Weighted Regression and Smoothing Scatterplots

Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models

us billion dollar weather and climate disasters in historical context including new county-level exposure, vulnerability and projected damage

Labor market impacts of covid-19 on hourly workers in small-and medium-sized businesses: Four facts from homebase data

Anomaly detection based on uncertainty fusion for univariate monitoring series

Time-series Extreme Event Forecasting with Neural Networks at Uber

STR: A Seasonal-Trend Decomposition Procedure Based on Regression

Terpenning. Stl: A seasonal-trend decomposition

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names