Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.prolog > #14468

Re: ILP is still dreaming of higher order (Was: Prolog Education Group clueless about the AI Boom?)

From Mild Shock <janburse@fastmail.fm>
Newsgroups comp.lang.prolog
Subject Re: ILP is still dreaming of higher order (Was: Prolog Education Group clueless about the AI Boom?)
Date 2025-03-08 00:00 +0100
Message-ID <vqftqi$16pcl$2@solani.org> (permalink)
References <vq4a7g$10843$1@solani.org> <vqftm2$16pcl$1@solani.org>

Show all headers | View raw


You are probably aiming at the decomposition
of autoencoder or transformer into an encoder
and decoder. Making the split up automatically

from a more general ILP framework.

 > The H is the bottleneck on purpose:
 >
 > relation(X, Y) :- encoder(X, H), decoder(H, Y).

Ok you missed the point. Lets assume for the
moment the H on purpose is not something that
happens accidential through a more general

learning algorithm. But it is a design feature
of how we want to learn. Can we incorporate
analogical reasoning, the parallelogram?

Yeah, relatively simple, just add more input
and output layers. The new parameter K indicates
how the representation was chosen:

relation(X, Y) :-
    similar(X, A, K),
    encoder(A, H),
    decoder(H, B),
    similar(Y, B, K).

Its again an autoencoder respectively transformer,
with a bigger latent space. Prominent additional input
layers that work here are convolutional neural networks.

Things like max pooling or self-attention pooling:

relation(X, Y) :-
    encoder2(X, J),
    decoder2(J, Y).

encoder2(X, [K|H]) :-
    similar(X, A, K),
    encoder(A, H),

decoder2([K|H], Y) :-
    decoder(H, B),
    similar(Y, B, K).

You can learn the decoder2/2 as a whole in your
autoencoder and transformer learning framework,
provided it can deal with many layers, i.e.

if it has deep learning techniques.

Mild Shock schrieb:
> The first deep learning breakthrough was
> AlexNet by Alex Krizhevsky, Ilya Sutskever
> and Geoffrey Hinton:
> 
>  > In 2011, Geoffrey Hinton started reaching out
>  > to colleagues about “What do I have to do to
>  > convince you that neural networks are the future?”
>  > https://en.wikipedia.org/wiki/AlexNet
> 
> Meanwhile ILP is still dreaming of higher order logic:
> 
>  > We pull it out of thin air. And the job that does
>  > is, indeed, that it breaks up relations into
>  > sub-relations or sub-routines, if you prefer.
> 
> You mean this here:
> 
>  > Background knowledge (Second Order)
>  > -----------------------------------
>  > (Chain) ∃.P,Q,R ∀.x,y,z: P(x,y)← Q(x,z),R(z,y)
>  >
>  > https://github.com/stassa/vanilla/tree/master/lib/poker
> 
> Thats too general, it doesn’t adress
> analogical reasoning.
> 
> Mild Shock schrieb:
>> Concerning this boring nonsense:
>>
>> https://book.simply-logical.space/src/text/2_part_ii/5.3.html#
>>
>> Funny idea that anybody would be interested just now in
>> the year 2025 in things like teaching breadth first
>> search versus depth first search, or even be “mystified”
>> by such stuff. Its extremly trivial stuff:
>>
>> Insert your favorite tree traversal pictures here.
>>
>> Its even not artificial intelligence neither has anything
>> to do with mathematical logic, rather belongs to computer
>> science and discrete mathematics which you have in
>> 1st year university
>>
>> courses, making it moot to call it “simply logical”. It
>> reminds me of the idea of teaching how wax candles work
>> to dumb down students, when just light bulbs have been
>> invented. If this is the outcome
>>
>> of the Prolog Education Group 2.0, then good night.
>>
> 

Back to comp.lang.prolog | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Prolog Education Group clueless about the AI Boom? Mild Shock <janburse@fastmail.fm> - 2025-03-03 14:18 +0100
  Re: Prolog Education Group clueless about the AI Boom? Mild Shock <janburse@fastmail.fm> - 2025-03-03 14:20 +0100
    Salary Templates if you "Grok" ML / AI [PhDs Negotiate Salaries] (Re: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-03-03 18:03 +0100
  ILP is still dreaming of higher order (Was: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-03-07 23:58 +0100
    Re: ILP is still dreaming of higher order (Was: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-03-08 00:00 +0100
  FYI: Philip Zucker’s Co-Egraphs (Was: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-08-12 18:37 +0200
    Using Hopcroft & Karp (HK) everywhere (Was: FYI: Philip Zucker’s Co-Egraphs) Mild Shock <janburse@fastmail.fm> - 2025-08-14 12:31 +0200
      Confusing "decidable problem" and "complete algorithm" (Was: Using Hopcroft & Karp (HK) everywhere) Mild Shock <janburse@fastmail.fm> - 2025-08-14 14:31 +0200
        DFA algorithms in a Python library (Re: Confusing "decidable problem" and "complete algorithm") Mild Shock <janburse@fastmail.fm> - 2025-08-14 15:14 +0200
      Static Variable Shunting in Dogelog Player (Was: Using Hopcroft & Karp (HK) everywhere) Mild Shock <janburse@fastmail.fm> - 2025-08-16 11:22 +0200
        Dynamic Variable Shunting in WebPL (Was: Static Variable Shunting in Dogelog Player) Mild Shock <janburse@fastmail.fm> - 2025-08-16 11:34 +0200
          SWI-Prolog is still the OG of GC (Was: Dynamic Variable Shunting in WebPL) Mild Shock <janburse@fastmail.fm> - 2025-08-16 11:46 +0200
            Cyclic Term Unification is Accounted (Was: SWI-Prolog is still the OG of GC) Mild Shock <janburse@fastmail.fm> - 2025-08-16 11:59 +0200
            WebPL is an interesting project (Was: SWI-Prolog is still the OG of GC) Mild Shock <janburse@fastmail.fm> - 2025-08-17 18:00 +0200
              Head to Head Race with Scryer Prolog (Was: WebPL is an interesting project) Mild Shock <janburse@fastmail.fm> - 2025-08-17 18:24 +0200
                Does Variable Age make Sense? [Prolog Unification] (Was: Head to Head Race with Scryer Prolog) Mild Shock <janburse@fastmail.fm> - 2026-02-10 12:05 +0100
                Type systems for non-deterministic concurrency [Amir Pnueli] (Was: Does Variable Age make Sense? [Prolog Unification]) Mild Shock <janburse@fastmail.fm> - 2026-07-20 12:20 +0200
                Robin Milners fickle gives non-determinism in practice [Parallel π-WAM] (Was: Type systems for non-deterministic concurrency [Amir Pnueli]) Mild Shock <janburse@fastmail.fm> - 2026-07-20 12:34 +0200
  Sleepy Joe and the Poor South (Local AI Emacs Mode) (Re: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-11-12 12:59 +0100
    Beyond Sleepy Joe around the World (Was: Sleepy Joe and the Poor South (Local AI Emacs Mode)) Mild Shock <janburse@fastmail.fm> - 2025-11-12 13:12 +0100
  GENESIS MISSION: Loosing it over Deep Pokets [Business wants A-Life] (Was: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2025-11-26 18:07 +0100
  Vanilla Prolog: semi-decidable =\= decidable (Re: Prolog Education Group clueless about the AI Boom?) Mild Shock <janburse@fastmail.fm> - 2026-03-14 20:40 +0100
    Its on the Internet, so it must be true? (Was: Vanilla Prolog: semi-decidable =\= decidable) Mild Shock <janburse@fastmail.fm> - 2026-03-16 11:01 +0100
    Hack + Computed Goto = Leightweigh C_OR (Was: Vanilla Prolog: semi-decidable =\= decidable) Mild Shock <janburse@fastmail.fm> - 2026-07-04 16:58 +0200
      The stack and choice points as an after match (Re: Hack + Computed Goto = Leightweigh C_OR) Mild Shock <janburse@fastmail.fm> - 2026-07-04 17:07 +0200
        Introduction to AI Accelerator Prolog [π-WAM of Dogelog] (Was: The stack and choice points as an after match) Mild Shock <janburse@fastmail.fm> - 2026-07-17 10:49 +0200
          Not praying to the god of lambda calculus [π beats α] (Was: Introduction to AI Accelerator Prolog [π-WAM of Dogelog]) Mild Shock <janburse@fastmail.fm> - 2026-07-17 11:13 +0200
          pi in pi-WAM refers to pi-calculus (Re: Introduction to AI Accelerator Prolog [π-WAM of Dogelog]) Mild Shock <janburse@fastmail.fm> - 2026-07-18 01:22 +0200
            Milners fickle() in pi-WAM [For fun and profit] (Re: pi in pi-WAM refers to pi-calculus (Re: Introduction to AI Accelerator Prolog [π-WAM of Dogelog]) Mild Shock <janburse@fastmail.fm> - 2026-07-18 01:49 +0200
              Robin Milners pi calculus is typeless (Re: Milners fickle() in pi-WAM [For fun and profit]) Mild Shock <janburse@fastmail.fm> - 2026-07-18 11:00 +0200
                Can the Church Turing hypotheses be refuted? [TLo @ FOM] (Re: Robin Milners pi calculus is typeless) Mild Shock <janburse@fastmail.fm> - 2026-07-18 11:10 +0200
          Accelerate Lean! From Theorem 3.11 to Corollary 3.12 [ZMC] (Was: Introduction to AI Accelerator Prolog [π-WAM of Dogelog]) Mild Shock <janburse@fastmail.fm> - 2026-07-19 16:06 +0200
            ANN: Library Pegg, for π-E-graphs [Not EXWM] (Was: Accelerate Lean! From Theorem 3.11 to Corollary 3.12 [ZMC]) Mild Shock <janburse@fastmail.fm> - 2026-07-19 16:21 +0200

csiph-web