Differentially private multivariate time series forecasting of aggregated human mobility with deep learning: Input or gradient perturbation? - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Article Dans Une Revue Neural Computing and Applications Année : 2022

Differentially private multivariate time series forecasting of aggregated human mobility with deep learning: Input or gradient perturbation?

Résumé

This paper investigates the problem of forecasting multivariate aggregated human mobility while preserving the privacy of the individuals concerned. Differential privacy, a state-of-the-art formal notion, has been used as the privacy guarantee in two different and independent steps when training deep learning models. On one hand, we considered gradient perturbation, which uses the differentially private stochastic gradient descent algorithm to guarantee the privacy of each time series sample in the learning stage. On the other hand, we considered input perturbation, which adds differential privacy guarantees in each sample of the series before applying any learning. We compared four state-of-the-art recurrent neural networks: Long ShortTerm Memory, Gated Recurrent Unit, and their Bidirectional architectures, i.e., Bidirectional-LSTM and Bidirectional-GRU. Extensive experiments were conducted with a real-world multivariate mobility dataset, which we published openly along with this paper. As shown in the results, differentially private deep learning models trained under gradient or input perturbation achieve nearly the same performance as non-private deep learning models, with loss in performance varying between 0.57% to 2.8%. The contribution of this paper is significant for those involved in urban planning and decision-making, providing a solution to the human mobility multivariate forecast problem through differentially private deep learning models.
Fichier principal
Vignette du fichier
2022_DPDL_Time_Series_Input_VS_Gradient.pdf (549.41 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03689723 , version 1 (07-06-2022)

Identifiants

Citer

Héber Hwang Arcolezi, Jean-François Couchot, Denis Renaud, Bechara Al Bouna, Xiaokui Xiao. Differentially private multivariate time series forecasting of aggregated human mobility with deep learning: Input or gradient perturbation?. Neural Computing and Applications, 2022, 34, pp.13355-13369. ⟨10.1007/s00521-022-07393-0⟩. ⟨hal-03689723⟩
208 Consultations
43 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More