Deep and Confident Prediction for Time Series at Uber

Abstract

Reliable uncertainty estimation for time series prediction is critical in many fields, including physics, biology, and manufacturing. At Uber, probabilistic time series forecasting is used for robust prediction of number of trips during special events, driver incentive allocation, as well as real-time anomaly detection across millions of metrics. Classical time series models are often used in conjunction with a probabilistic formulation for uncertainty estimation. However, such models are hard to tune, scale, and add exogenous variables to. Motivated by the recent resurgence of Long Short Term Memory networks, we propose a novel end-to-end Bayesian deep model that provides time series prediction along with uncertainty estimation. We provide detailed experiments of the proposed solution on completed trips data, and successfully apply it to large-scale time series anomaly detection at Uber.

arXiv commentTo appear in DSBDA-2017 @ ICDM'17

References in corpus (15)

• Auto-Encoding Variational Bayes
• Explaining and Harnessing Adversarial Examples
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• Unsupervised Learning of Video Representations using LSTMs
• Weight Uncertainty in Neural Networks
• A Theoretically Grounded Application of Dropout in Recurrent Neural Networks
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Bayesian Recurrent Neural Networks
• Variational Bayesian Inference with Stochastic Search
• Dropout Inference in Bayesian Neural Networks with Alpha-divergences
• Nonlinear Systems Identification Using Deep Dynamic Neural Networks
• Modeling asymptotically independent spatial extremes based on Laplace random fields
• Concrete Dropout
• Black-box $\alpha$-divergence Minimization

Cited by in corpus (56)

• Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
• Deep Learning for Anomaly Detection: A Survey
• Deep Learning on Traffic Prediction: Methods, Analysis and Future Directions
• Deep Learning for Time Series Anomaly Detection: A Survey
• Deep Learning for Time Series Forecasting: Tutorial and Literature Survey
• What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use
• Combining time-series and textual data for taxi demand prediction in event areas: a deep learning approach
• Probabilistic Charging Power Forecast of EVCS: Reinforcement Learning Assisted Deep Learning Approach
• A Survey on Epistemic (Model) Uncertainty in Supervised Learning: Recent Advances and Applications
• Recurrent Neural Networks for Time Series Forecasting
• A Joint Model for IT Operation Series Prediction and Anomaly Detection
• Survey of Dropout Methods for Deep Neural Networks
• Predictive Monitoring with Logic-Calibrated Uncertainty for Cyber-Physical Systems
• Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting
• Uncertainty-Aware Attention for Reliable Interpretation and Prediction
• Uncertainty estimation of pedestrian future trajectory using Bayesian approximation
• Foundations of Sequence-to-Sequence Modeling for Time Series
• Probabilistic Deep Learning to Quantify Uncertainty in Air Quality Forecasting
• On Last-Layer Algorithms for Classification: Decoupling Representation from Uncertainty Estimation
• Fast Transient Simulation of High-Speed Channels Using Recurrent Neural Network
• On Hallucination and Predictive Uncertainty in Conditional Language Generation
• Prediction Uncertainty Estimation for Hate Speech Classification
• Radflow: A Recurrent, Aggregated, and Decomposable Model for Networks of Time Series
• Leveraging World Events to Predict E-Commerce Consumer Demand under Anomaly
• A Deep Learning based Detection Method for Combined Integrity-Availability Cyber Attacks in Power System
• Probabilistic Time Series Forecasting with Implicit Quantile Networks
• Exploring Bayesian Deep Learning for Urgent Instructor Intervention Need in MOOC Forums
• Improving Back-Translation with Uncertainty-based Confidence Estimation
• RNN-based counterfactual prediction, with an application to homestead policy and public schooling
• Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation
• Bayesian Methods for Semi-supervised Text Annotation
• Simultaneously Reconciled Quantile Forecasting of Hierarchically Related Time Series
• Uncertainty-aware Attention Graph Neural Network for Defending Adversarial Attacks
• SAFARI: Safe and Active Robot Imitation Learning with Imagination
• Applying SVGD to Bayesian Neural Networks for Cyclical Time-Series Prediction and Inference
• Self-Supervised Log Parsing
• The Monte Carlo Transformer: a stochastic self-attention model for sequence prediction
• LSTM for Model-Based Anomaly Detection in Cyber-Physical Systems
• Prediction intervals for Deep Neural Networks
• Deep Distribution Regression
• Online Learning Models for Vehicle Usage Prediction During COVID-19
• Uncertainty Intervals for Graph-based Spatio-Temporal Traffic Prediction
• Exploring Context Generalizability in Citywide Crowd Mobility Prediction: An Analytic Framework and Benchmark
• Machine Learning Algorithms for Financial Asset Price Forecasting
• Harnessing value from data science in business: ensuring explainability and fairness of solutions
• Quantifying Uncertainty in Deep Spatiotemporal Forecasting
• Accurate and Reliable Forecasting using Stochastic Differential Equations
• Simulation of Open Quantum Dynamics with Bootstrap-Based Long Short-Term Memory Recurrent Neural Network
• Nonparametric Online Learning Using Lipschitz Regularized Deep Neural Networks
• Recoding latent sentence representations -- Dynamic gradient-based activation modification in RNNs
• A new approach to forecast service parts demand by integrating user preferences into multi-objective optimization
• Spatiotemporal Pattern Mining for Nowcasting Extreme Earthquakes in Southern California
• Uncertainty Surrogates for Deep Learning
• Switching Scheme: A Novel Approach for Handling Incremental Concept Drift in Real-World Data Sets
• Predicting Requests in Large-Scale Online P2P Ridesharing
• Short-term Load Forecasting with Deep Residual Networks