Philippe Decottignies

[Thierry Bayon @MARKETOR_IT]

#IntelligenceArtificielle : 13 tendances lourdes pour 2018, selon une étude CB Insights à lire absolument... - #iA #AI https://t.co/BrJPGoqd86

[Reza Zadeh @Reza_Zadeh]

A fantastic use of Fourier Transforms to draw a painting. These days you might hear it called a "Generative Fourier Neural Network" ;). By @jagarikin.

[RE•WORK Deep Learning @reworkdl]

How CapsNets can overcome some shortcomings of CNNs, including requiring less training data, preserving image details, and handling ambiguity: https://t.co/P9oNuqEHrM

[RE•WORK Deep Learning @reworkdl]

Something to do with your weekend! 5 Free Exhaustive And Comprehensive Resources To Learn Deep Learning From Scratch: https://t.co/VGqcYRC8Kt

[Reza Zadeh @Reza_Zadeh]

Overview of why Deep Reinforcement Learning needs work:
- Sample Inefficient
- Simpler methods outperform
- Reward Function hard to design
- Unstable / irreproducible

Similar to Deep Learning 10 years ago. Some amazing successes but not plug-n-play, yet. https://t.co/97jnblYjrN

[Dustin Tran @dustinvtran]

"The shift from AI being a research domain to it increasingly becoming a research + engineering domain, is a strong signal that we're not in a bubble this time." +100. Systems research advances science and is immediately useful :) https://t.co/kLtu8iXxsa

[Hello Future @Orange_Future]

Que va changer l'Intelligence Artificielle dans notre quotidien : évolution ou révolution ? https://t.co/b2iv7gwptj

[Philippe Decottignies @phdecottignies]

Configuration du Movidius Neural Compute Stick (79€ de la taille d’une clé USB) connecté à un Raspberry Pi 3 (34€) => l’IA a portée des objets connectés, un vaste champ des possibles vient d’apparaître 😄

[Philippe Decottignies @phdecottignies]

@defts @AUCHAN_France Ça craint en effet. Bien vu David