Second meetup in 2020 in the heart of Kreuzkölln, two great talks and of course great food, drinks and lots of fun.
PLEASE NOTE: We are setting the attendee limit purposefully very high. Seats at the meetup will be on a first come first serve basis. We have space for about 100 people. The meetup will be located on the second floor.
Filip Danieluk: Trusting your pet project
This talk is about two things, a pet project of mine and eXplAInability. The pet project's aim? Drive a car in a video game setting. The undertaken approach was to train a CV model that classifies each frame to a specific command - turning left, right, or moving forward.
To explain the decisions made by the model I used LRP (Layer-wise Relevance Propagation). I will briefly explain how LRP works and how it compares with more popular algorithms, especially LIME.
Filip is a graduate of Warsaw University of Technology, Faculty of Power Engineering. He taught himself Python 4 years ago and has been part of PyData Warsaw Conference committee since its first edition - 3 years ago. He is employed by CBRE as Data Scientist working with a variety of data problems in the real estate sector.
Bharathi Srinivasan: Uncertainty in Recommender Systems
Bayesian deep learning is gaining popularity as they help us reason about uncertainty. This application shows how these methods can be extended to recommender systems.
Bharathi is a Machine Learning Engineer working on recommender systems at Comtravo. She is an engineer by background and a Master's graduate from Humboldt University of Berlin. Her interests in machine learning include probabilistic programming, Bayesian methods of deep learning and causal inference.
Lightning talk slots are still open.
Be tuned for some exciting news that we will share with you!