Re:First-Class Automatic Differentiation in Swift
by bmc7505 on Monday October 22, 2018 @02:25AM
(#18272588)
The idea is that in the not-too-distant-future, differentiable programming will become much more mainstream. c.f. https://medium.com/@karpathy/software-2-0-a64152b37c35
Re:First-Class Automatic Differentiation in Swift
by saagarjha on Monday October 22, 2018 @02:35AM
(#18272614)
Ok, but to use terms from that article, Swift is still very much "Software 1.0". It's a general-purpose programming language meant to be written by people. Sure, it's good that people are trying to extend it to fit their specific needs, but I don't think that Swift (or any commonly used language today) is anywhere near being "good" for these types of tasks.
Re:First-Class Automatic Differentiation in Swift
by wool_gather on Monday October 22, 2018 @08:35AM
(#18274094)
> Neural networks [...] represent the beginning of a fundamental shift in how we write software.
That's an...interesting statement, considering that neural networks are nearly as old as software itself. The original Pitts-McCulloch paper was published in 1943. Active use of them has certainly been ongoing longer than that blog author has been alive.
Re:First-Class Automatic Differentiation in Swift
by tree_of_item on Monday October 22, 2018 @08:41AM
(#18274122)
And yet it's true, since they haven't been used in the way they are now. There was no DeepMind in 1943. Computer science stuff usually takes a few decades to mature.
Personally, I feel like it would be unwise to merge in all of TensorFlow's requirements into Swift. Many features, such as the ones you mentioned, are actually useful to outside of the context of machine learning, so it does actually make sense to make them generally available. But compiler support for something as specific as @differentiable seems like an anti-goal for Swift.