Skip to Main content Skip to Navigation
Conference papers

Probabilistic Time Series Forecasting with Structured Shape and Temporal Diversity

Abstract : Probabilistic forecasting consists in predicting a distribution of possible future outcomes. In this paper, we address this problem for non-stationary time series, which is very challenging yet crucially important. We introduce the STRIPE model for representing structured diversity based on shape and time features, ensuring both probable predictions while being sharp and accurate. STRIPE is agnostic to the forecasting model, and we equip it with a diversification mechanism relying on determinantal point processes (DPP). We introduce two DPP kernels for modeling diverse trajectories in terms of shape and time, which are both differentiable and proved to be positive semi-definite. To have an explicit control on the diversity structure, we also design an iterative sampling mechanism to disentangle shape and time representations in the latent space. Experiments carried out on synthetic datasets show that STRIPE significantly outperforms baseline methods for representing diversity, while maintaining accuracy of the forecasting model. We also highlight the relevance of the iterative sampling scheme and the importance to use different criteria for measuring quality and diversity. Finally, experiments on real datasets illustrate that STRIPE is able to outperform state-of-the-art probabilistic forecasting approaches in the best sample prediction.
Document type :
Conference papers
Complete list of metadatas

Cited literature [65 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02992358
Contributor : Vincent Le Guen <>
Submitted on : Friday, November 6, 2020 - 1:10:22 PM
Last modification on : Friday, December 4, 2020 - 10:08:15 AM

File

STRIPE_NeurIPS20.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02992358, version 1

Collections

Citation

Vincent Le Guen, Nicolas Thome. Probabilistic Time Series Forecasting with Structured Shape and Temporal Diversity. NeurIPS 2020, Dec 2020, Vancouver, Canada. ⟨hal-02992358⟩

Share

Metrics

Record views

46

Files downloads

22