Thursday, August 8, 2019

Nash equilibria and hardness

Back when there was more of a logic focus, it seemed easier to argue for limitations since one could use terms like undecidable-ness. From my experience in KBE, we found enough events that puzzled to talk 'vertigo' (yes, lost in the fog). From an operational sense, we used a smart 'man-in-the-loop' to cut ourselves out of intractable situations (got 'er done).

Then, Bayes came along. He is not recent; it is just that computation finally caught up. I used him for a paper five decades ago and heard tsk-tsk. But, then, they laughed at Leontief, then, too. There were other tricks that have evolved. Systems became (much) more powerful. Lots of cheap energy was around and about. Too, seemingly, notions of constraint were left on the side of the road.

Data existence ballooned, though there are many issues left to discuss. That allowed a whole industry to grow (data science). AI/ML/DL took advantage of the situation. Games were won that had not been within grasp before.
Linguistics advanced, too. Computer-generated text rose like the sun. Yet, to this old-guy's eyes, we're dealing with a situation not unlike the million monkeys and a novel. Uncountable numbers of them creating?

The stuff does not resonate. But, I'm talking talents not recognized, yet. So, that's not even on the table.

To me, numeric abilities allowed a peanut-buttering without this being obvious. They force convergence. It looked, to hear the hype, that we're now all in a rosy time (colored glasses, for sure). But, again, the old guy will point to 12 years ago when the risk experts were saying, piece of cake - there will never be another downturn (on the eve of the very thing).

What gives? We need to discuss. So, Aviad shows hardness. There are regular discussions at Stanford and elsewhere.
To the old guy's mind, it's like a re-opening of what was known. Or taking it back off of the shelf. We'll have to rephrase some stuff. But, there are limits. And, there are things that AI needs to know dealing with structure and more. It's nice to see Nash re-evaluated.

Used for a Quora question: With Nash equilibria being shown as intractable, ought we consider that there are numeric limits to AI that are being ignored?

Remarks: Modified: 08/08/2019

08/08/2019 --

Thursday, May 9, 2019

Test by tragedy

Unfortunately, we get this, especially when things get complicated.

Ethics might be a way to bring decisions related to two recent events back to the table.

Remarks: Modified: 08/08/2019

08/08/2019 -- Let's just say this, an overly-complicated procedure came out with an unstable design that would have been seen in the older days. However, the process was changed and cut too far. The issue got by and was given to the compute folk to handle. Tsk.  

Sunday, December 30, 2018

Summary, 2018

Prior years:  20112012201320142015, 2016, 2017 (missing), 2018.

Remarks: Modified: 12/30/2018

12/30/2018 --

Sunday, November 18, 2018


So, we're past what is called the mid-term (something like that) election. So, the blues took the House from the reds. That ought to dampen some of the whining. We will see. The Fed has started to quit the flaying of the savers. Some savings rates are up to 2% and more.

We never whined here. No, it has always been righteous indignation. Like the prophet of old, we could not see much but thing going crazy.

However, now is the time for reasonable folks to step forward, and we intend to do so.

Of late, our creative juices have been spent on two avenues: Thomas Gardner Society, Inc. site and blog; Quora, the indescribable, so to speak.

As we look at the influence of social media on society, we have to get back into that realm's need for truth engineering. We intend to use wind-speed social media as an anchor.

Remarks:  Modified: 11/18/2018

11/18/2018 --

Saturday, October 13, 2018

Overview of AI

The gist of this post is a recap, somewhat, that is being done briefly. I have been working in Quora, for the most part, but intend to get back to this blog.

The recent CACM of the ACM published an overview of AI by the head of CompSci at UCLA. His view resonated real strongly with me. So, I want to provide a quick look at some of the key items.

First, the article: Human-level intelligence or animal-like abilities?  Author: Adnan Darwiche.

Adnan starts by summarizing some of the recent successes (if you can call them that) that got so much attention in the press. The computer (AI driven) winning in Go. The computer learning how to play a game and beating expert humans. The list goes on. And, the basis is the neural net approach that is not new. Marvin Minsky wrote of these one-half a century ago.

But, too, people got upset. Some are believing in the coming singularity. I say, worry about the current crapularity. Oh, the robots are coming. Just like there was a warning 200 years ago about the invasion of the Brits who wanted to spank the free Americans.

Adnan mentions some of the mainstream areas of AI that have been ignored in the rush after the numeric paradise that is being given us by machine learning. He mentions "knowledge representation, symbolic reasoning, and planning" in short. Adnan was there when the AI winter came about from the mania of the '80s. He was student.

I would tell him that knowledge based engineering came out of that thrust and went fine for decades doing great things. I can (and will) point to writings about this. However, let's continue with Adnan's look from his lofty position.

He reminds us that AI had a reasoning approach, from the beginning. The recent work? No, it's numeric processing, pure and simple. Now, he mentions that particular neural nets (NNs, of which there are many variety) are doing some type of representation of a model. Given that, we can understand these supposed black-boxes with the proper approach. He uses two concepts: model-based and function-based. The former was (is) a huge part of AI, in general. You know, the latter was seen in engineering, too.

So, what we are talking is that the tried-and-true of science and engineering somehow got lost in the rush after the search for the magic function (oh yes, there ain't none). The main argument for the NNs is the advancements that we see in "sophisticated statistical and optimization techniques" that do seem to have amazing results.

But, as Adnan said, it seems to be that we're chasing what animals already do well. And, not doing as well as they. The article provides sufficient examples. But, consider this: no animal can make a computer.

Too, there seems to be a huge divide between the true believers here (chasing after money, power, fame, what have you) and those more scientifically oriented. Besides, the whole of the atmosphere has been polluted with out-of-control hype. Adnan uses 'bullied by success' (let's talk some type of innoculation, shall we?).

What is real here? Well, Adnan has the answer. He says that we're dealing with 'function engineering' where one goal is to approximate cognitive functions. There are more. In fact, Adnan's position is strongly influenced by Judea Pearl who recently wrote on the subject (The Book of Why: The New Science of Cause and Effect).

To me, Adnan is talking the old middle-out technique that has been the reality of complicated systems all along. We could think of the NNs as supporting bottom up. But, be aware that there are more areas of technology awaiting attention. The top-down would be somewhat assisted by the traditional approaches. Albeit, we are really talking the human expert rising to the challenge. Per usual.

However, NNs might just be universal as some have claimed from the beginning (not Marvin as he was a critic to his death) in some simulation sense. Guess what? Physics and reality might be closely mimicked by our computationalist's approaches, however being is there irregardless. The 'crap' from extensive computer work (and there are whole bunches to denote) grows. Unfortunately, that whole bit of accumulating residue gets no (or little) attention.

Throughout the paper, Adnan quotes anonymous sources. There is good reason to not get into the cross-hairs at this time. What is really at stake? Well, truth engineering has been dealing with these issues from its inception (my work of the past two decades). So, our pace has (might have) been that of a snail, yet it's tied into other than gaming. Oh yes, we have a whole lot of work to do in reviewing the just past decade and one-half.

Remarks: Modified: 10/16/2018

10/16/2018 -- As Adnan suggested, since there is little scientific thinking going on, I see that we have raging imagination causing the 'running amok' atmosphere. And, we forget the past. This little study has major importance: Carnegie Mellon is Saving Old Software from Oblivion. Those who know have been concerned that experiments (heavily reliant upon computers) cannot be repeated due to technology due to several factors. One is the flash push (wow factor) versus sustainable science. 

Friday, February 16, 2018


We have a web-site associated with this work, albeit we have not really used it much: It will be re-organized. Also, we will be working with the Thomas Gardner Society, Inc. in areas that overlap. To be brief, as one looks at the discussions about truth and related around the world, there is a growing influence from social media. Quora is an example.

But, discussions are all over the map. AI, especially ML/DL, is a sidetrack with robotics seen as a new life form. So much is going on that ignores the importance of people to the issue of reality and truth. Yes, people are part of the process in so many ways.

However, too, the U. S. stands as a symbol, for a whole lot of things. That will be discussed. But, one fact is that the history of the U.S. needs to be re-examined. History is told by the victor, it is said. I say, history is told by people who like to talk and write.

When you look at history, we have many contributors whose story is told by others, and not very well. We have chance to look at that with Thomas Gardner. We, basically, know his from his children and the descendants thereof. Yet, Thomas was right there from the beginning (1623) of the U.S. through the establishment of Massachusetts which, by the way, subsumed the Plymouth colony.

All through the history of the U.S., there are threads from people from those early arrivals who carried on the activity related to the establishment and evolution of the country. Generation by generation, they were there for the good times and the heart ache. Yet, history glosses over that.

Truth, though, pertains to people more than to anything. Arguable, perhaps, but, then not.

Remarks: Modified: 02/16/2018

02/16/2018 --

Wednesday, November 29, 2017

Modern ways

Not posting here does not mean that there are not things going on. For the past few months, I have been writing on Quora, with the intent of getting back here. In fact, the whole aspect of truth engineering has been brought out there. The intent is to update this blog in the near future.

For one thing, I ran into an interest in Common Lisp on the part of the younger set. There are several manifestations in use. One called clojure has been used for both front-end and back-end work. What is suggest to me is that we can start to talk a workbench approach.

Aside: There is a lot to discuss, but Lisp has been involved with AI from the beginning. Here is some information related to why it is so good: What-did-Alan-Kay-mean-by-Lisp-is-the-greatest-single-programming-language-ever-designed

If one listens to some AI/DS/ML/DL (that is, Artificial intelligence, data science, machine learning, deep learning) practitioners, one sense can an out-of-control situation. Gobs of time, money, space get gobbled running against suspect data, one might say. As, this information is wide and not very deep. So, who is deep? In other words, where is the science?

Too, there is a general malaise with regard to not understanding. That is, the computer comes up with something, and the humans cannot provide anything that allows comprehension to look at it and make a judgment. No, the reaction is to think the the computer is out of our league.

That might be true in several senses, but it is not in terms of truth (hence, we need to engineer this).

Remarks: Modified: 11/29/2017

11/29/2017 --