Folks who are into this may also like the "Type Safe Neural Networks in Haskell" https://blog.jle.im/entry/practical-dependent-types-in-haske...

I'm also starting work on a set of bindings to libdarknet for Idris with similar properties.

Note that the article has a comment by Yann LeCun (hopefully it's not an impersonator).

You may also be interested in Differentiable Neural Computers:

- https://deepmind.com/blog/differentiable-neural-computers/

- https://github.com/deepmind/dnc

> Note that the article has a comment by Yann LeCun (hopefully it's not an impersonator).

i wouldnt be surprised if it was lecun. colah's illustrations on nonlinear transformations have made it into several lecun papers, including the following [nature review]( https://www.nature.com/articles/nature14539).

Following the link from Yann LeCun's comment in the parent post one can find this neat library: DiffSharp http://diffsharp.github.io/DiffSharp/

Discussed at the time: https://news.ycombinator.com/item?id=10165716.

Some related work also include:

- "Strongly-Typed Recurrent Neural Networks" http://proceedings.mlr.press/v48/balduzzi16.pdf

- Principled Approaches to Deep Learning workshop http://padl.ws/ (maybe this is in line with the meta-point Colah's paper)

- Haskell accelerate library https://github.com/AccelerateHS/accelerate/. Not deep learning per se but perhaps some of the ideas are applicable

This covers some of same topics plus the rapidly expanding #s of frameworks, SIMD /SIMT backends etc https://julialang.org/blog/2017/12/ml&pl

20 pg tutorial on why RNN's are tricky https://arxiv.org/abs/1801.01078

Many mentions of Python, none of Numba, which does a good job of jitting Python with LLVM.

It seems unlikely though that an entire modeling community could rally behind a single language or framework, given all the possibilities, many of which are commercially oriented. But one I’ve used recently with a lot of flexibility is Loopy

https://documen.tician.de/loopy/

There's also this take on differentiating datatypes: https://codewords.recurse.com/issues/three/algebra-and-calcu...

Reading this makes a lot of the operations in colah's article feel more intuitive. (To me, at least. I'm no expert here.)

The examples are odd because he doesn't incorporate any notion of differentiability.

So a Generating RNN is not quite like foldr, since foldr has no notion of differentiability.

One needs to show examples that pulls in some kind of automatic-differentiation capability.

When you pass a differentiable function into fold -- or most higher order functions, for that matter -- you get a function that is differentiable on everything but a measure zero set.

The mechanics of how you compute the derivatives are separate from this. Obviously, the efficient way is to use backprop (reverse mode AD), as we always do in deep learning. But you could also use discrete derivative approximations. The point is that the resulting function is differentiable, which is independent of how you compute the derivatives.