XPost: rec.puzzles, alt.usage.english
From: HenHanna@NewsGrouper
Jeff Barnett posted:
> On 10/25/2025 4:23 PM, HenHanna@NewsGrouper wrote:
> >
> > HenHanna posted:
> >
> >> On Wed, 26 Mar 2025 17:15:58 +0000, Carl G. wrote:
> >>
> >>> On 3/25/2025 9:59 PM, HenHanna wrote:
> >>>> (What are these sequences?) --- (For each line, What is the
next
> >>>> letter?)
> >>>>
> >>>>
> >>>> 1. H, H, L, B, B, C, N, ?
> >>>>
> >>>> 2. D, P, N, G, C, M, M, S, N, ?
> >>>>
> >>>> 3. U, Q, P, P, Q, L, O, M, ? --------** (See below)
> >>>>
> >>>> 4. U, D, T, C, C, S, S, ?
> >>>
> >>> 1. "O" (It's elementary)
> >>
> >>
> >>
> >> omg... you guys are so good at this.
> >>
> >>
> >> I'm very sorry for the error.... The 3rd (mystery) sequence is:
> >>
> >> 3. Q, P, Q, P, Q, L, O, M, ?
> >>
> >>
> >>
> >>
> >> ----- it's still very hard. almost impossible.
> >
> >
> > Within a few days, I'll try these on my AI.
>
> FYI Paul Abrahams did his PhD dissertation at MIT in the early 1960s;
> it's purpose was to solve sequence problems and was considered AI-like
> at the time. When he grew up, he was a professor at Courant Institute, a
> President for awhile of the ACM (Association for Computer Machinery),
> and a devoted fan of Figure 8 Stock Car Racing.
>
> His dissertation handled such things as stop sequences on the New York
> subway lines and many puzzles such as the above in addition to more
> math-based problems. For the latter, there is now a days
> The On-Line Encyclopedia of Integer Sequences! at
> https://oeis.org/
> --
> Jeff Barnett
>
>
My AI can't solve these, but has good suggestions.
My AI can't solve the Square sliding (clockwise) puzzle.
-------- what shapes do the points a,b,c form?
_______________________________
Thank you... thats good to know... I'll look that up.
I remember another famous project from MIT (BASEBALL analogy)
The MIT AI Lab project from the 1970s related to baseball analogies
is likely connected to the early work on AI problem-solving and analogy
reasoning, where baseball was used as a domain to model questions and
reasoning problems.
One known program from that era was called "Baseball," developed around 1960
by Bert F. Green Jr. and colleagues at MIT AI Lab.
It was able to answer constrained questions about baseball games like "Who did
the Red Sox lose to on July 5, 1960?" This early question-answering program
modeled a small, well-defined domain (a baseball season) to demonstrate AI's
capacity for
understanding and reasoning with data. It was a pioneering example of using AI
to process natural language queries in a real-world context.
This project is a notable early milestone at MIT AI Lab linking baseball
analogies to artificial intelligence, showing how domain-specific knowledge
(here baseball) was leveraged to advance research on AI understanding and
problem-solving in the 1970s.
_______________________
The MIT AI Lab project called "BASEBALL," developed around the 1960s-1970s by
Bert F. Green Jr. and colleagues, focused on analogy-making and question
answering within the constrained domain of professional baseball games from a
season's data.
The kind of analogies BASEBALL could make or understand involved reasoning
about baseball events and statistics by using patterns and logical
relationships. For example, it could answer questions like "Who did the Red
Sox lose to on July 5, 1960?" by
relating facts about teams, dates, and game outcomes. The program used
structured knowledge and analogy-like inference to navigate the domain and
retrieve correct answers, simulating a reasoning process over a small,
well-defined dataset.
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
