Path: ...!news.nobody.at!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Paul Rubin <no.email@nospam.invalid>
Newsgroups: comp.lang.forth
Subject: Re: Differentiable Forth
Date: Tue, 16 Jul 2024 16:12:50 -0700
Organization: A noiseless patient Spider
Lines: 16
Message-ID: <87h6cpgcct.fsf@nightsong.com>
References: <e9689378e46f06961fe4fe43b47dfd3b@www.novabbs.com>
	<v766m4$3hipk$1@news.xmission.com>
	<f806c73034c83af94d0ce3d857075241@www.novabbs.com>
	<v76ris$3htrd$1@news.xmission.com>
	<b83ad41e25a48ccb727b94f427665073@www.novabbs.com>
MIME-Version: 1.0
Content-Type: text/plain
Injection-Date: Wed, 17 Jul 2024 01:12:51 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="1ad292c86056a3c03a94a525affde3a8";
	logging-data="1566286"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19q4fkgSF0Lt00mLiaGs8Zw"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux)
Cancel-Lock: sha1:R4LkmVVA0XPuv/mAvxqfG/zYXSE=
	sha1:MY9zBYTCgeTIONDVcmCR/j4IYoM=
Bytes: 1718

mhx@iae.nl (mhx) writes:
> It's the wrong type of differentiation again. AI-people are
> overloading our vocabulary, redefining our words, and overflowing our
> stacks.

The automatic differentiation there looks like what you were talking
about, built into the language sort of like some languages have built in
complex numbers.

Here is a 2005 article about AD that you might like:

http://blog.sigfpe.com/2005/07/automatic-differentiation.html

It mentions at the end that you can approximate this method using
complex numbers.  Just pretend that the imaginary part is nilpotent, I
guess.