Deep Factors with Gaussian Processes for Forecasting

2018/11/30 by Danielle C. Maddix, Yuyang Wang, Maddix, Danielle C. +3 · 5 citations
Computer Science · Mathematics · #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Time Series Analysis and Forecasting #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1812.00098

Third workshop on Bayesian Deep Learning (NeurIPS 2018), Montreal, Canada

arxiv created 2018/11/30 · openalex publication_date 2018/11/30 · arxiv updated 2018/12/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

A large collection of time series poses significant challenges for classical and neural forecasting approaches. Classical time series models fail to fit data well and to scale to large problems, but succeed at providing uncertainty estimates. The converse is true for deep neural networks. In this paper, we propose a hybrid model that incorporates the benefits of both approaches. Our new method is data-driven and scalable via a latent, global, deep component. It also handles uncertainty through a local classical Gaussian Process model. Our experiments demonstrate that our method obtains higher accuracy than state-of-the-art methods.

Citations

• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Auto-Encoding Variational Bayes
• MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
• A Multi-Horizon Quantile Recurrent Forecaster
• GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration
• DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks

Cited by

• Structured Noise Modeling for Enhanced Time-Series Forecasting
• Probabilistic Forecasting with Temporal Convolutional Neural Network
• Deep Factors for Forecasting
• CANet: ChronoAdaptive Network for Enhanced Long-Term Time Series Forecasting under Non-Stationarity
• FISformer: Replacing Self-Attention with a Fuzzy Inference System in Transformer Models for Time Series Forecasting

Related

• Deep Factors for Forecasting
• Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting
• Tailored Architectures for Time Series Forecasting: Evaluating Deep Learning Models on Gaussian Process-Generated Data
• A Comprehensive Survey of Deep Learning for Time Series Forecasting: Architectural Diversity and Open Challenges
• Time Series Forecasting With Deep Learning: A Survey