• 16
 
20TH ANNIVERSARY 09/12 APRIL 2019 MARRIOTT CONFERENCE CENTER PARIS


Tuesday 09 April AI NET Tutorial Track 3

    Conference Room Auditorium Level -1

  • Registration and welcome coffee from 07.30
  • Start of the TUTORIAL 14.00
  • Exhibition open from 09.00 to 19.30
  • Seated Lunch: 12.30
  • End of TUTORIAL 17.00


12.30
Seated Lunch

        Tutorial presented by Orange Labs Representants
14.00
Deep Learning, Supervised Learning, LSTM, CNN, Timeseries: «A Day in the Life»

This tutorial covers end to end machine learning including deep learning techniques applied to Network Data.

The goal is to share a methodology that sketches all the steps from raw data transformation and visualization to model generation and application including performance evaluation. The tutorial is based on real network data.
  • Raw network data and their transformation into a tabular format

  • Visualization and network data analysis (distribution, correlation, etc.)

  • Preprocessing of network data and which function to consider depending on the data shape (e.g. scaling, normalization, outlier removal, differentiation, etc.)

  • A Network data forecasting is presented to illustrate how important and complex is to configure a neural network. Several variations of the hyper parameters are shown to assess their impact on the forecasting of network timeseries

  • A Network data anomaly detection with deep learning is presented to illustrate a typical classification operation and to benchmark several deep learning architectures to clarify the impact on the classification problem
17.00
End of the Tutorial