TF&DL October : TensorFlow 4k


This month we celebrate our MeetUp Group attaining its 4,000th member! Among TensorFlow MeetUp groups worldwide, Singapore is right up there at/near the top in terms of total number of people. Of course, on a per-capita basis we COMPLETELY DOMINATE!

Planned Talks :

"Training Models at Scale with TPUs: Donuts, Pods and Slices" - Sam Witteveen
As models and datasets get bigger, the need for accelerators that can massively speed up your training is crucial. In this talk, Sam will discuss the current state of how to train models with TPUs, the changes in code needed in TensorFlow 2.0 to run on TPUs and what is coming in the near future.

"Explanations for School Science Questions" - Yew Ken Chia
The TextGraphs Workshop at EMNLP (a conference being held in Hong Kong at the beginning of November) had a Shared Task competition. This went beyond the simple "Answer the multiple-choice question correctly" task that AllenAI made waves with earlier this month. The task involved coming up with the correct chain of reasoning : i.e. multi-step explanations. Ken will discuss the work that went into the team's paper.

"TensorFlow Probability and Bayesian Hyperparameter Optimisation" - Martin Andrews
With AutoML is all the rage, Martin will show how the TF Probability package can be used to model Gaussian Processes, and how that is relevant to Hyperparameter Optimisation.


Talks will start at 7:00pm and end at around 9:00pm, at which point people normally come up to the front for a bit of a chat with each other, and the speakers.


As always, we're actively looking for more speakers for future events - both '30 minutes long-form', and lightning talks. For the lightning talks, we welcome folks to come and talk about something cool they've done with TensorFlow and/or Deep Learning for 5-10mins (so, if you have slides, then #max=10). We believe that the key ingredient for the success of a Lightning Talk is simply the cool/interesting factor. It doesn't matter whether you're an expert or and enthusiastic beginner: Given the responses we have had, we're sure there are lots of people who would be interested to hear what you've been playing with. Please suggest yourself here :