Published on

Learn to Learn to Learn (Singapore)

Presentation
Authors

At last the Singapore AI Day arrived!

Our speaker line-up included :

  • A Recap from TensorFlow Summit - Sorawit Saengkyongam, Google Developer Expert Machine Learning

  • Machine Learning in JavaScript with TensorFlow.js - Virot Chiraphadhanakul, Google Developer Expert Machine Learning

  • Parallelising Machine Learing using R - Chris Auld, Principal Technical Evangelist Manager - Southeast Asia at Microsoft

  • Building AI Products: From Paper to Prototype to Production - Sam Witteveen, Google Developer Expert Machine Learning

  • Learn to Learn to Learn - Martin Andrews, Google Developer Expert Machine Learning (me)

  • Deep learning in life sciences - Allen Day, Developer Advocate, Google

  • Reinforcement Learning for Connect4 - Hardeep Arora, Head AI/Advanced Analytics, Accenture

  • Data Science at GoJek

The talk I gave was an updated version of the FOSSASIA workshop I gave previously, but without having people follow along with their own Virtual Machines, due to time construints.

I chose to give greater weight to the last three of the four sections :

  • Machine Learning in TensorFlow Playground
  • Using Keras model zoo for the pre-trained NASNet model, and doing transfer learning there
  • Going through the Reptile meta-learning algorithm (implemented in PyTorch)
  • Showing how TensorFlow Eager mode can accomplish the same, line-by-line

As usual, all the source for the deeplearning-workshop notebooks is available on GitHub - if you have questions on the software, please leave an 'issue' there.

Presentation Screenshot

If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.

Presentation Content Example

The presentation was kindly recorded by Engineers.sg.