horde-ad-0.3.0.0: CREDITS.md

Credits go to

Mikolaj Konarski, the original author of the library

Conal Elliott, talking beautifully about the Essence of Automatic Differentiation

Simon Peyton Jones, who sketched in "Provably correct, asymptotically efficient, higher-order reverse-mode automatic differentiation" talk at https://www.youtube.com/watch?v=qGkgLtpseMs algorithms that this Haskell codebase implements

Faustyna Krawiec, Neel Krishnaswami, Tom Ellis, Andrew Fitzgibbon, Richard Eisenberg, the remaining authors of the paper "Provably correct, asymptotically efficient, higher-order reverse-mode automatic differentiation" (https://dl.acm.org/doi/10.1145/3498710) that describes and proves correct a comprehensive AD formalism, including some ingenious algorithmic ideas this Haskell codebase implements

Oleg Grenrus, whose https://hackage.haskell.org/package/overloaded contains inspiring examples related to AD

Edward Kmett, author of the archetype Haskell https://hackage.haskell.org/package/ad library

Justin Le, author of the https://hackage.haskell.org/package/backprop library and the amazing articles at https://blog.jle.im

Lennart Augustsson, author of the incredibly clever shapely typed Haskell tensor library https://hackage.haskell.org/package/orthotope

Tom Smeding, who influenced the major implementation decisions of horde-ad and solved the hardest technical problems

and other benefactors and contributors, in chronological order:
Stanisław Findeisen, Cale Gibbard, Well-Typed LLP


Authors of files `t10k-images-idx3-ubyte.gz`, `t10k-labels-idx1-ubyte.gz`, `train-images-idx3-ubyte.gz`, `train-labels-idx1-ubyte.gz` from http://yann.lecun.com/exdb/mnist, copied into this repo into directory `samplesData/`, copyright (according to https://keras.io/api/datasets/mnist) Yann LeCun and Corinna Cortes for the MNIST dataset, which is a derivative work from original NIST datasets, copyright by their authors and made available under the terms of the Creative Commons Attribution-Share Alike 3.0 license.