MLT Workshop: Convolution Operation

Machine Learning Tokyo is organizing a new deep learning workshop, an Introduction to Convolution Operation @Rakuten. We'll learn theoretical concepts underlying convolutional neural networks as well as some of the most important CNN architectures, including small interactive implementation blocks.

This workshop is free of charge and limited to 60 seats. For Rakuten security reasons, every participant needs to register with full name and contact details here: https://goo.gl/forms/2RwGwvTmrTd8Lm893

• • SCHEDULE - -

Part I :

• The Theory Behind Convolutional Operations
• Mathematical and Visual Representation
• Convolutions in Computer Vision
• Snippet Implementation

LUNCH (huge thank you to Rakuten for sponsoring lunch!)

Part II:

• Convolutions in Deep Architectures

• 3x3 vs 11x11
  ------ AlexNet vs VGG

• 1x1 conv and 1x1 vs FC
  ------ GoogleNet/Inception

• Depthwise (separable) conv
  ------ MobileNet

• Channel shuffling
  ------ ShuffleNet vs MobileNet

• Residuals in conv-nets
  ------ ResNet vs ResNexT

• Connections in conv-nets
  ------ ResNet vs DenseNet

• • INSTRUCTORS - -
    Alisher Abdulkhaev, Machine Learning engineer @PKSHA Technology
    Mustafa Yagmur, Machine Learning engineer @Weathernews
    Dimitris Katsios, Machine Learning engineer @LPixel
• • MINIMUM REQUIREMENTS - -
    Intermediate Python and Machine Learning
    Basic Deep Learning (intuition, implementation)
• • REGISTER NOW - -
    https://goo.gl/forms/2RwGwvTmrTd8Lm893

#####################################################
A big THANK YOU to Rakuten for providing not only the venue for this workshop, but also lunch for all participants!

Rakuten is one of the world's leading e-commerce and internet companies. Their vision is to drive disruptive innovation and engage knowledge, creativity and passion from around the world to achieve ambitious goals and help build communities in which people can pursue their dreams and live in happiness. https://global.rakuten.com/corp/about/philosophy/
####################################################################