Thomas KurbielinTowards Data ScienceSpatial Transformer Networks — BackpropagationA Self-Contained IntroductionOct 12, 20211Oct 12, 20211
Thomas KurbielinTowards Data ScienceSpatial Transformer NetworksA Self-Contained IntroductionSep 27, 2021Sep 27, 2021
Thomas KurbielinTowards Data ScienceSpatial Transformer Networks Tutorial, Part 2 — Bilinear InterpolationA Self-Contained IntroductionSep 13, 2021Sep 13, 2021
Thomas KurbielinTowards Data ScienceSpatial Transformer Tutorial, Part 1 — Forward and Reverse MappingA Self-Contained IntroductionAug 30, 2021Aug 30, 2021
Thomas KurbielinTowards Data ScienceTransformer Networks: A mathematical explanation why scaling the dot products leads to more stable…How a small detail can make a huge differenceApr 28, 20215Apr 28, 20215
Thomas KurbielinTowards Data ScienceDerivative of the Softmax Function and the Categorical Cross-Entropy LossA simple and quick derivationApr 22, 202111Apr 22, 202111
Thomas KurbielinTowards Data ScienceAleatory Overfitting vs. Epistemic OverfittingApproaching the two reasons why your model is not able to generalize wellDec 20, 20201Dec 20, 20201
Thomas KurbielinTowards Data ScienceDeriving the Backpropagation Equations from Scratch (Part 2)Gaining more insight into how neural networks are trainedNov 23, 20204Nov 23, 20204
Thomas KurbielinTowards Data ScienceDrawing the Transformer Network from Scratch (Part 1)Getting a mental model of the Transformer in a playful wayNov 15, 20207Nov 15, 20207
Thomas KurbielinTowards Data ScienceDeriving the Backpropagation Equations from Scratch (Part 1)Gaining more insight into how neural networks are trainedNov 8, 20202Nov 8, 20202