Catch up on the latest talks from Munich Datageeks — fresh insights, cutting-edge tech, and real-world data stories from top minds in the field. Stay curious and stay inspired!
Learn more about the hottest topic's in data and AI.
Sampling-based Bayesian deep learning outperforms traditional approximations in both uncertainty quantification and predictive accuracy. Modern algorithms like microcanonical Langevin Monte Carlo enable efficient sampling of networks with millions of parameters.
Framework for evaluating LLMs in production classification systems using three components: golden datasets for ground truth benchmarking, LLMs as judge for semantic validation, and human feedback for real-world performance tracking.
A 10+ year journey of building and continuously evolving a smart cat flap system - from basic Raspberry Pi sensors to machine learning, Home Assistant integration, and reverse-engineered hardware. A tale of persistence, over-engineering, and why some projects never get boring.
Deep reinforcement learning AI achieves competitive esports lap times in Need for Speed Most Wanted using reverse-engineered game memory and a custom training environment, running on 10-year-old consumer hardware in just 24 hours of training.
Google engineer Christoph Mittendorf demonstrates AI use cases at Munich Datageeks: Like Lederhosen trend prediction using search data, running form analysis against Eliud Kipchoge, and Genie 3 open world models for gaming and robotics applications.
