Published on

XLNet (Singapore AI Day)


This year's Singapore AI Day was prompted by the arrival of 3 people from Google Brain (Mountain View, CA) - and we also managed to get Sirinart Tangruamsub (a GDE based in Thailand) to come and talk about her work with Bayesian Linear Regression.

Our speaker line-up for the 4-hour (standing-room only) event was :

  • "Introduction to Keras in TensorFlow 2.0" - Paige Bailey, Google Brain

  • "Deep Learning on Graphs for Conversational AI" - Sam Witteveen, Red Dragon AI

  • "Which image should we show? Neural Linear Bandit for Image Selection" - Sirinart Tangruamsub, Agoda

  • "TFX" - Robert Crowe, Google Brain

  • "XLNet - The Latest in language models" - Martin Andrews, Red Dragon AI

  • "Swift for TensorFlow" - Paige Bailey, Google Brain

  • "TensorFlow Lite: On-Device ML and the Model Optimization Toolkit" - Jason Zaman, Light

The talk I gave was the 'cutting edge' one for the event, with the XLNet paper only having been published 10 days before the event...

Half of my talk was introducing Transformer models generically, the next one third was about the innovations in XLNet, and the final third (yes, I overran a little) was a series of '1-minute papers' describing lots of cool stuff that people are doing with Transformers.

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