Published on June 28, 2022–Updated on June 28, 2022
Dates
on the July 1, 2022
from 10:30am to 12:00pm
Location
Inria Sophia Antipolis Méditerranée
Program
10:30 - 10:45
Prof. Ludovic Dibiaggio
Introduction to the OTESIA institute
10:45 - 11:15
Daniel Inzunza (Inria, ACUMES)
PINNs approach for traffic model calibration
Abstract: The increasing amount of data generated by smartphones and IoT devices motivated the development of Federated Learning (FL), a framework for on-device collaborative and privacy-preserving training of machine learning models. Heterogeneity is a core challenge in Federated Learning: devices with higher availability are likely to participate more often during the training, thus introducing a bias in the learned model. Our theoretical analysis points out a tradeoff between 1) letting everybody participate leading to a faster convergence towards a surrogate of the objective function; 2) correcting the bias resulting in a slower convergence to the true objective. In our work, we propose an energy-efficient solution to alleviate the bias without neglecting the convergence speed.
11:15 - 11:45
Angelo Rodio (Inria, NEO)
Resource-aware Federated Learning
Abstract: At the macroscopic scale, traffic is usually represented as a fluid flowing through the road network. To characterize the evolution of the system, aggregate quantities such as the flow Q = Q(t, x) (it veh/h) or the density rho = rho(t, x) (it veh/km) of vehicles are defined. The challenge is to be able to understand, reproduce and anticipate the evolution of density and flow in space and time based on both this mathematical modeling and traffic data. Existing physical traffic flow models, such as Lighthill-Whitham-Richards (LWR) models, which can only capture real-world traffic dynamics to a limited extent, result in low-quality estimation and require massive data for accurate and generalizable estimation. To solve this problem, we use Physics-Informed Neural Networks (PINNs) techniques which have been shown to date to be an efficient numerical tool that provides solutions to partial differential equations (PDEs), even though, theoretically, they have limited ability to solve problems with continuous solutions. Particularly, we focus on the LWR model with observed loop detector data, using traffic density as the traffic variables. We show the advantages and disadvantages of PINNs technique for solving (with loop detector data) LWR physical traffic flow models, and suggest strategies for dealing with the problems that arise when we use neural networks.
When browsing Université Côte d'Azur website and Université Côte d'Azur components websites by profile ("I am" menu), informations may be saved in a "Cookie" file installed by Université Côte d'Azur on your computer, tablet or mobile phone. This Cookie file contains informations, such as a unique identifier, the name of the portal, and the chosen profile. This Cookie file is read by its transmitter. During its 12-month validity period, it allows to recognize your terminal and to propose the chosen profile as your default home page.
You have accepted the deposit of profile information cookies in your navigator.
You have declined the deposit of profile information cookies in your navigator.
"Do Not Track" is enabled in your browser. No profiles information will be collected.
Cookies de mesure d 'audiences
This website uses Google Analytics. By clicking on "I accept" or by navigatin on it, you authorize us to deposit a cookie for audience measurements purposes.
You have accepted the deposit of audience measurement cookies in your navigator.
You have declined the deposit of audience measurement cookies in your navigator.
"Do Not Track" is enabled in your browser. No navigation statistics will be collected.