Ce soir: Paris Machine Learning Meetup #13 Season 3; Data augmentation methods in Deep Learning, Neural Turing Machines, Manga Recommendation

All the previous archives of the 40+ meetups are here.The hangout will be live:

We will be hosted by DojoEvents. Refreshements and food will be offered by dotAI.io. Thank you to them for their support. There will be 2 free tickets to the next dotAI.io to win for people at the meetup.

 Follow our Twitter account, and / or use the following hashtag #MLParis during the meetup.

PROGRAM :

+ Franck Bardol, Igor Carron, Introduction

+ Pitches (2mins):

• Blue Frog Robotics, Jean-Michel Mourier, CTO, presentation
• LightOn, Igor Carron, CEO, presentation

+ Meilleur Data Scientist de France (epilogue d'hier), Jeremy Atia , Romain Ayres , Christophe Bourguignat, Stéphane Fenia, Fabien Vauchelles

+ Vinay Kumar (arya.ai)
Role of Data augmentation methods in Deep Learning

With CNN architecture, Deep Learning models can predict/classify the data with better accuracy when compared any other methods. But instead of creating a new network architectures for new problem, data augmentation methods can help in using the same architecture and address to many tasks. In this talk, we will discuss about the various data augmentations we can use in Deep Learning and develop a model using arya.ai. 

+ Tristan Deleu (snips.ai)
Learning Algorithms (and more) with Neural Turing Machines

The idea of adding an external memory to neural networks has been increasingly popular over the past 2 years. In this talk, I will present you one of the earliest example of this family of models: the Neural Turing Machines. I will show you how they work, their early applications to algorithm learning as well as and recent developments for language understanding.

Jill-Jênn Vie (mangaki.fr)
Mangaki, recommandation de mangas open source

La plupart des sites marchands sont capables de nous recommander des produits correspondant à nos goûts. Mais comment cela fonctionne ?
Nous décrirons deux algorithmes permettant d'effectuer cela, ainsi qu'un test adaptatif de bienvenue pour l'élicitation des préférences.
 

Join the CompressiveSensing subreddit or the Google+ Community or the Facebook page and post there !
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.