Data-Driven Assumptions
Max and Tim delve into the fundamental debate of whether machine learning can be purely data-driven or if prior assumptions are necessary. Max emphasizes the importance of balancing inductive biases with data availability, highlighting the core trade-offs in machine learning. The discussion challenges the notion of a blank slate approach and explores the role of assumptions in creating smooth mappings for effective learning.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#036 - Max Welling: Quantum, Manifolds & Symmetries in ML
Related Questions