✳flâneur — a map of the web's best reading
(KANs part 1) An introduction to B-splines
rohangautam.github.io · 4,991 words · saved by 1 readers
Let's talk about B-splines, a popular approximation and interpolation curve, and the workhorse of the Kolmogorov-Arnold network
(KANs part 1) An introduction to B-splines Skip to main content (KANs part 1) An introduction to B-splines 18 June 2024 machine learning , splines , KAN In this blog series, I'll be going through the technical details involved in understanding Kolmogorov-Arnold Networks - a new type of machine learning architecture which has gotten significant attention lately due to it's several advantages over MLPs(Multi Layer Perceptrons), such as interpretability and avoiding catastrophic forgetting to name a few. Table of Contents: In part one of this series, let's build our understanding from the ground
Explore this link on the map →saved by
related reading
- B-spline - Wikipediaen.wikipedia.org
- bspline - B-spline interpolation with Python - Stack Overflowstackoverflow.com
- B-spline Curves: Important Propertiespages.mtu.edu
- intuition - Splines: relationship of knots, degree and degrees of freedom - Cross Validatedstats.stackexchange.com
- A Primer on Bézier Curvespomax.github.io
- KANarxiv.org
- Kolmogorov–Arnold representation theorem - Wikipediaen.wikipedia.org
- [2404.19756] KAN: Kolmogorov-Arnold Networksarxiv.org
- Smoothstep - Wikipediaen.wikipedia.org
- Napkin.pdfvenhance.github.io
- The Engineering behind Figma’s Vector Networksalexharri.com
- Cubic Interpolation of Quaternionstheorangeduck.com