Interpretable Deep Learning with R, Keras and LIME

Date: 
Tuesday, October 16, 2018 - 18:00
Source: 
R-Ladies Coding Club (London)
Attendees: 
58
City: 
London

Registration via Eventbrite: https://www.eventbrite.co.uk/e/interpretable-deep-learning-with-r-lime-and-keras-tickets-50118369392

Remember, without an Eventbrite ticket your registration won't be valid!

ABOUT THE SPEAKER

R-Ladies London have a great pleasure to host Dr Shirin Glander, a data scientist at codecentric in Munster, Germany. If that wasn't enough, she's also an R-blogger, data enthusiast, organiser of M√ľnsteR, animal lover, ballroom and latin dance teacher! You can read her award-winning blog here: https://www.shirin-glander.de/

Shirin will talk about using R to build an interpretable imagine recognition models.

ABOUT THE TALK

Keras is a high-level open-source deep learning framework that by default works on top of TensorFlow. Keras is minimalistic, efficient and highly flexible because it works with a modular layer system to define, compile and fit neural networks. It has been written in Python but can also be used from within R. Because the underlying backend can be changed from TensorFlow to Theano and CNTK (with more options being developed right now) it is designed to be framework-independent. Models can be trained on CPU or GPU, locally or in the cloud. Shirin will show an example of how to build an image classifier with Keras. We'll be using a convolutional neural net to classify fruits in images. But that's not all! We not only want to judge our black-box model based on accuracy and loss measures - we want to get a better understanding of how the model works. We will use an algorithm called LIME (local interpretable model-agnostic explanations) to find out what part of the different test images contributed most strongly to the classification that was made by our model. Shirin will introduce LIME and explain how it works. And finally, she will show how to apply LIME to the image classifier we built before, as well as to a pre-trained Imagenet model.

MEETUP AGENDA

18.00 - 19.00: Doors open and networking.
19.00 - 19.15: Intro in the Data Science at Farfetch
19.15 - 20.30: Shirin's talk + questions
~20.30 - ~21.00: networking

ACKNOWLEDGEMENTS
We would like to say thanks to our sponsor; tech recruitment agency RedCat Digital (https://www.redcat-digital.com/), and to our hosts global ecommerce platform Farfetch (https://www.farfetch.com/uk/). Together with them, we are pleased to promote diversity and create a community of minority genders in the data space!

The Bower (Warehouse, 3rd floor)

207-211 Old Street, EC1V 9NR