SophI.A Summit à Sophia Antipolis les 7, 8 et 9 novembre 2018, est le premier événement à portée internationale, dédié à l’Intelligence Artificielle, organisé par un collectif associant collectivités, associations, université et entreprises du territoire.
Son programme propose un contenu varié qui s’adresse à différents publics : académique, économique et citoyen.
Au programme de ces 3 jours :
- Des conférences scientifiques de haut niveau réunissant le monde académique et celui de l’entreprise pour des regards croisés autour de diverses thématiques d’actualité ;
- Des événements participatifs de type Master Classes et Start-ups pitchs, ou interactifs tels que des poster session, pour mettre en pratique et se challenger ;
- Deux soirées de sensibilisation à destination du grand public ;
- Un spectacle digital #SoftLove.
MASTER CLASSES 
La journée du Mercredi 7 novembre, Telecom Valley organise des Master Classes.
Deux sessions de 3 Master Classes en parallèle, de 2h30 chacune, l’une le matin (début à 10h) et l’autre l’après-midi (début à 14h).
Le contenu des 2 sessions est similaire le matin & l’après-midi.
Vous pouvez choisir 1 sujet de Master Class le matin et choisir un autre sujet l’après-midi parmi les 3 sujets suivants :
Infuse AI @scale in your business with IBM Watson Studio
Jean-Luc Collet, Cognitive Solution Architect at IBM Global Industry Solution Center Nice-Paris
Philippe Gregoire, IBM EcoSystem Advocacy Hub Nice/Europe
Greger Ottosson, PhD, AI Strategy & Digital Business Automation Lead at IBM France Lab Sophia-Antipolis
Alain Chabrier, PhD, Decision Optimization for Data Science, Technical & Integration Lead at IBM France Lab Sophia-Antipolis
Greger Ottosson
En savoir plus
Whether you’re a Business user or a Data scientist, in this Master Class you will discover hands-on how to apply & infuse AI in your business use cases – end to end. You will use IBM Watson Studio to first ingest, cleanse and analyze data, and then build, train and deploy Machine Learning models. You will also move beyond the predictive power of machine learning and see how the « prescriptive analytics » technology in IBM Decision Optimization can generate optimal decisions for your business.
Equally important, we will cover how to address some of the core business challenges with machine learning, such as model robustness, accuracy and bias. This category also includes how to explain decisions and how business rules technology complements machine learning for business-critical decisions.
Prerequisite
This Soph.I.A. 2018 masterclass comprises Hands-On Labs which will be executed by the masterclass participants on IBM Cloud instances, using a web browser running on their own laptops, PC; Mac or true Linux. The IBM Watson Studio environment leverages IBM Cloud infrastructure and accounts. The creation of an IBM Cloud account is required before the masterclass, following the instructions hereafter.
Have an e-mail address available, one which has never been used to setup an IBM Cloud account
From a web browser, navigate to https://console.bluemix.net/registration/?cm_mmc=SophIA18
Enter the requested details
10:00 AM - 1:00 PM
2:00 PM - 5:00 PM
SAP Predictive Analytics
Pierre-Emmanuel POUMELLEC, SAP Developer Expert
Sofiane LOUNICI, Developer Internship
Aymen MOUELHI
En savoir plus
SAP Predictive Analytics, enables data analysts to easily carry out high-quality, robust predictions with little effort.
Tasks that that typically require deep expertise and high effort in conventional data mining environments are handled automatically.
For instance SAP Predictive Analytics takes care of:
- Handling of missing values and outliers
- Binning of continuous variables
- Grouping of categories.
- Multicollinearity
- Column selection
- Model Selection
The user interface is designed for high usability.
Thanks to a high performance engine, SAP Predictive Analytics can handle very large data volumes – models can be created on thousands of columns. No prior column selection has to be carried out. Any column that might possibly have an impact on the score can be included. Once the model has been created, the user can look into the model to understand its findings and workings. »
Prerequisite
No expert scripting knowledge is required.
The installation of the tool SAP Predictive Analytics is required before the class :
It has to be downloaded as Free trial account from https://www.sap.com/cmp/ft/crm-xm15-dwn-an001/index.html.
10:00 AM - 1:00 PM
2:00 PM - 5:00 PM
Hyperbase, a web toolbox for text mining and deep learning - by BCL, Université Côte d'Azur
Laurent VANNI, Ingénieur d’étude à Université Côte d’Azur
Damon MAYAFFRE, Researcher CNRS at UMR 7320 : Bases, Corpus, Langage, Université Côte d’Azur
En savoir plus
Hyperbase is a web platform for statistical analysis of textual data. The user interface allows users to load their corpus of text and access a tool box for statistical analysis of their data. From simple word frequency analysis to deep learning, Hyperbase allows linguistic analysis of and clustering of data according to the weight of each text or word.
The objective of this master class is to take this platform in hand and test the different tools available through examples of corpus available online.
Prerequisite: To fully understand the examples, a good understanding of French is desirable.
10:00 AM - 1:00 PM
2:00 PM - 5:00 PM
Le nombre de places est limité.