Adaptive Basis Functions
Tim and Thomas discuss how neural networks adaptively place basis functions in regions with high error, focusing on areas where the underlying function changes the most. They explore the difference between data density and information density, highlighting how neural networks diverge from classical techniques in capturing essential information.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#69 DR. THOMAS LUX - Interpolation of Sparse High-Dimensional Data
Related Questions