Deep Probabilistic Forecasting

Deep probabilistic forecasting aims to generate not only point predictions but also probability distributions representing the uncertainty inherent in time series predictions, improving decision-making under uncertainty. Current research focuses on enhancing accuracy and uncertainty quantification through novel training methods that account for error autocorrelation and incorporating prior knowledge like calendar-driven periodicities, often employing neural networks such as autoregressive models, temporal convolutional networks, and transformers. These advancements are significant for various applications, including air quality forecasting, financial modeling, and resource management, where understanding prediction uncertainty is crucial for reliable decision-making. Furthermore, there's a growing emphasis on developing more interpretable models to increase trust and facilitate practical adoption in sensitive domains.

Papers

December 22, 2023

SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecasting
Shane Bergsma, Timothy Zeyl, Lei Guo
Time Series Autoregressive Model Long Sequence Probabilistic Forecast Sequence Model Autoregressive Generative Model Deep Probabilistic Forecasting

May 26, 2023

Better Batch for Deep Probabilistic Time Series Forecasting
Vincent Zhihao Zheng, Seongjin Choi, Lijun Sun
Probabilistic Forecast Neural Forecasting Deep Probabilistic Forecasting Trend Correlation

February 6, 2023

ProbPNN: Enhancing Deep Probabilistic Forecasting with Statistical Information
Benedikt Heidrich, Kaleb Phipps, Oliver Neumann, Marian Turowski, Ralf Mikut, Veit Hagenmeyer
Probabilistic Forecast Statistical Query Periodicity Detection Deep Probabilistic Forecasting

January 5, 2023

DANLIP: Deep Autoregressive Networks for Locally Interpretable Probabilistic Forecasting
Ozan Ozyegen, Juyoung Wang, Mucahit Cevik
Human Prediction Inherent Interpretability Probabilistic Time Series Forecasting Interpretable Forecast Deep Probabilistic Forecasting

February 27, 2022

Taming the Long Tail of Deep Probabilistic Forecasting
Jedrzej Kozerawski, Mayank Sharan, Rose Yu
Probabilistic Forecast Heavy Tailed Long Tail Pareto Optimality Deep Probabilistic Forecasting Kurtosis Concentration

December 6, 2021

Parameter Efficient Deep Probabilistic Forecasting
Olivier Sprangers, Sebastian Schelter, Maarten de Rijke
Temporal Convolutional Network Transformer Based Approach Probabilistic Time Series Forecasting Deep Probabilistic Forecasting

December 5, 2021

Probabilistic Deep Learning to Quantify Uncertainty in Air Quality Forecasting
Abdulmajid Murad, Frank Alexander Kraemer, Kerstin Bach, Gavin Taylor
Uncertainty Quantification Model Uncertainty Predictive Uncertainty Air Quality Forecasting Probabilistic Deep Learning State of the Art Uncertainty Deep Probabilistic Forecasting