Getting started with JAX
Simon Pressler

Getting Started with JAX! Hands-on tips to overcome your first hurdles.

Honey, I broke the PyTorch model >.< - Debugging custom PyTorch models in a structured manner
Clara Hoffmann

Honey, I broke the Pytorch model >.< No problem! In this talk, we'll build a toolbox to debug our models and prevent this from happening again -all by leveraging DL logic, synthetic data and pytest. Let's make our models unbreakable <3

Leveraging Compound Sparsity to Achieve the Fastest Inference Performance on CPUs: "Why GPU Clusters Don't Need to Go Brrr"
Damian Bogunowicz

Fun fact: you can remove 90% of a neural network's weights without losing much accuracy! With model sparsity, you can even run these networks on your CPU with GPU-level performance. Learn about compound sparsity (pruning, quantization, knowledge distillation) for faster inference

Teaching Neural Networks a Sense of Geometry
Jens Agerberg

By taking neural networks back to the school bench and teaching them some elements of geometry and topology we can build algorithms that can reason about the shape of data. This is the promise of the emerging field of Topological Data Analysis (TDA) which we will introduce!