Posts Tagged ‘nks’

An interesting approach to knowledge mentioned in Stephen Wolfram’s blog:

But what about all the actual knowledge that we as humans have accumulated?

A lot of it is now on the web—in billions of pages of text. And with search engines, we can very efficiently search for specific terms and phrases in that text.

But we can’t compute from that. And in effect, we can only answer questions that have been literally asked before. We can look things up, but we can’t figure anything new out.

Let’s see where this goes!

Read Full Post »

The NKS summer school archive site is live.  I figured it would be best for me to wait until that was done before I attempted to post my project or write too much about other’s projects.

You can check out a summary of what I did personally on the Wolfram site.

Project Title 
Perturbing Turing Machines

Perturbations to elementary cellular automata have been investigated thoroughly. Under a certain level of perturbation, there are slight changes to local patterns but the automata tend to recover globally. The more complicated rules show greater disturbance but still can tolerate perturbations. This study considers similar perturbations to Turing machines.

Do Turing machines exhibit similar behavior?

“And the reason this is important is that in any real experiment, there are inevitably perturbations on the system one is looking at” (NKS p. 324). We must account for the effects of perturbations to draw any connections between these simple constructs and their natural counterparts.

Every project done there was interesting.  I encourage readers to check them all out.  Some projects are more abstract than others and all were a good launch pad for further research or immediate practical use.

I do have to call out Ben Rapoport’s project on neuronal computations.  It was beautiful in many ways.

Read Full Post »

Rather than expend energy writing my own general overview of what the heck just happened at summer school I’ll just link to this wrap up from the Wolfram team.

Sure, I’ll have far more details in upcoming posts, though most of those details will involve actual math, code, projects and implications and less about the school experience.


Read Full Post »

If the universe (our experience, our lives, our physical reality) weren’t complex (unpredictable, undecidable) what would it be?

This is not rhetorical question.

It is not easy either.

Can you imagine an alternative?

It would be useful if we could so we can go look for evidence of the thing you imagine.  Why would we do this?  The growing research in “complexity” and “complex systems” make some assumptions based on irreducibility, computation equivalence and so forth that suggest less complex things are not capable of universal computation, and in some sense, the ability to evolve ever more interesting/complex things, like our universe.

Read Full Post »

As I prep for a summer in Vermont to study NKS and automata, I’m starting to build research and project concepts.  My focus, as it stands now, is to some how take concepts from behaviorism (schedules of reinforcement, operants, rewards and punishers) and use automata to study them computationally.  This is not trivial nor is their any indication yet that it will be valuable.  

I’m attempting to mash the two lines of inquiry because there really isn’t an accurate nor reliable mathematical foundation in behaviorism even though the experimental and explanatory power of behaviorism is substantial and proven.

IF, and it’s a big IF, cellular automata and a computational (as opposed to a partial differential equation set) can model the concepts in behaviorism we will have a very exciting line of research to chase down.  Modeling and researching complex human behaviors (those with lots of overlapping and interacting schedules and complex environments) has been impossible experimentally and mathematically – only the most basic of behavior is possible to study and it usually has to be isolated to the point where it looses the environment it so richly interacts with.  If we can devise cellular automata capable of showing operant conditioning in ever more realistic environments, we could set up very complicated models without all the laboratory fixings….and so much more.

Note that I am not studying Social Behavior or Social Dynamics or Swarms – not in the typical “what you read on the blogs” sense.  A lot of work has been done in that area even with automata.  The study of individual behavior (how a particular individual responds and learns) needs more research.  I believe that a more thorough understanding of individual behavior will lead to stronger more robust social behavior models – as social behavior is emergent from individual behavior.

Anywho… to whet yer whistle read some fun stuff from Alastar Hewitt on mathematical reinforcement learning and CA



Read Full Post »

After a prodding from a pal, I decided to apply to the Wolfram Science NKS Summer School. NKS is a fun and exciting research area and Mathematica rocks. I have done some preliminary research in trying to merge automata to experimental analysis of behavior. Three solid weeks of study and interaction with others could do wonders to furthering those efforts! Heck, three weeks doing anything different always does wonders!

Below I present the thought process in applying to NKS Summer School. I’m not sure it’s interesting to anyone, but I am trying more and more to catalogue the web of contingencies involved in my own behavior.

John said, Go!, let’s go check out the app process. I got that same email, hmmm, I’ll check it out again. I saw the application form, looked simple. I read a few blog posts from the event last year. I swivele in my chair to see how accessible my NKS book was… oh, it is right there. I flip it open and read the preface and chapter 1. I look at the clock. Oh, shoot, I really need to finish this code for tomorrow’s 11am meeting in Santa Monica. Oh, shoot, this is so cool. I’ll read a bit more. Oh, look at this… when I picked this book up 4 years ago this all seemed so crazy hard and really he spends an awful lot of time and paper on repeating the same points about automata behavior. Gee the good stuff is left to an appendix after page 1100. OOo, he’s talkinga bout space and time and automata. That’s kinda fun and a different approach, oddly know equations or mathematica symbolics to mathematically meld them. lemme flip ahead…. check email. check clock. check blog stats. Crap it’s 2:30. and if i’m going to apply it has to be within 45 minutes or I’ll never do it. better take some notes. where was that stuff on responsibility and behavior? casual networks, that’s good stuff and interesting in that it meshes well with john’s visualizations. skinner, wonder what their opinion on skinner is. the phrasing in the book is somewhat dismissive of non computational sciences and of behaviorism. that’s usually the case. also their site is fairly devoid of social sciences and anything that goes beyond low level biology and physics. oh well, if i were to do this i’d want to mash behavior models to this. not sure you can even do that in 3 weeks. talking to wolfram himself would be cool. i should ask dani about this. yeah, but then I’ll question it. the app doesn’t lock you in. dump it in and see what happens. vermont for 3 weeks. hmmm, that’s cool. ok, how do I couch my response. they also want a resume. gotta dump that out from the blog and format it. should i submit as openoffice doc so they know i’m a tech. hey, i gotta download the latest release of mathematica, i got a note about that a week or so ago. that’s a big download, going to take awhile. alright, i have these notes. now I have to put together some indication of research interest. how to show detail in reading NKS? and what are they looking for… people who spent a lifetime working in it or is this a reach out… attract newbies who have solid interest? is commercial background a hinderance? is my commercial resume good enough, should it be reworked. cool, I went to uchicago. ENTP, wonder if they will get that. my resume is fun but certainly is an odd read. oh well, more time on that later. alright i’ve got these statements. ugh, lots of errors. this is why you don’t write in a text editor. takes too long to clean it all up. alright, i’ll correct the basics, read it outloud and then go. crap, running out of time. just get it in. fill it out… i bet this will crap out on upload and i’ll have to do it again. hahaha. oh, cool it worked. alright quick shoot it to john. ah crap, there’s a wrong tense or two and a mistyped word in the basic statements. not a lot I can do now. they have the blog and my resume. if i’m interesting as a personality and can twist NKS in new ways, maybe they will be into it. Let me quick confirm there’s no one who proposed some of my thinking… one dude did a social sciences lecture. crap requires mathematica to view. damnit, redoing linux messes up my flow. i should just do this on my laptop. based on the title and his bio doesn’t look much different than most social sciences. everything else typical bio, phys and math. alright, now i have to finish this code….first a vault soda and some cheez its.

Here are my original application answers to their open questions.

Tell us about your interest in Summer School?

NKS offers a new model for researching and applying an experimental analysis of behavior (empirical behaviorism). Popular sociology and psychology (and even some of the cognitive disciplines) rely far too much on statistical analysis and, perhaps, less accurate probabilistic models. The mathematical methods presented by NKS provides a structure that closely aligns with schedules of reinforcement we observe and the concept of selection by consequences.An intensive session in NKS should aid in my application of simple automata like models applied to the study of human behavior. In particular, I hope to research and complete a project based on social networks and the ever increasing merging of our offline and online lives.

My professional work in new media remains too “business oriented” to get to the root of what’s going on with human behavior and our interactions with information, the world and each other. Yet, I’ve had enough deep exposure to massive schedules of reinforcement, deep troves of user data and human computer interaction within companies like eHarmony, Yahoo!, Maxim, and Business.com to know that we now have the empirical data and computing capability to map extreme casual networks and model out possible simple rules generating the complexity of social behavior.

That said, I’m under no illusion that we’re close to fully mapping out or predicting human behavior. This summer session would aid in generating discussion and likely producing further avenues of research and applications of NKS and Mathematica.

In how much detail have your read NKS?

  I have completed NKS once in totality and referenced in regularly over the last 3.5 years. It has served as a research launching point for me into evolutionary programming, Dennet’s work, computation vs. statistical approaches and data compression issues in modeling. (among many other areas…) I maintain this cheatsheet of questions and NKS pointers on my desk for reference as I explore new media, experimental analysis of behavior, and social networks.

chapter 9
casual networks and schedules of reinforcement

want to explore the multi pathway of our existence. selection by consequences in behavior might just be simplistic automata like rules based on schedules of reinforcement…. purely human level… what does that even mean?

page 516 is right to the heart of my questioning. “given only a causal network, what can one say about the evolution history?” that’s human behavior and how we typically represent it in the media. We report on a limited view of the casual network and have a real difficulty discussing the explicitly evolution history.

page 518 “This has the consequence that effects in such systems can spread only at a limited rate, as manifest for example in a maximum slope for the edges of patterns…” human behavior seems to have this quality as well, which we ob Chapter 10 on perception is tied directly to human behavior. how we absorb new stimuli and mash it to discriminant stimuli. probabilities of behavior… the part on statistical analysis. Determinism in human behavior is a fundamental discussion point. Free Will.

page 596 is key point of research here….slight variation in the rules can produce dramatically different behavior and traditional statistical analysis (what is most done in sociology and psychology) cannot help identify changes in the casual networks and/or underlying rulesets “for it has almost always been assumed that to emulate in any generality a process as sophisticated as human thinking would necessarily require an extremely complicated system” That’s precisely the assumption behaviorism goes after….

NKS might be the mathematical backing to unwind the thinking.

bottom of 629 contains an important disclaimer… “the system would almost certainly have to have built up a human-like base of experience.” This is worth testing. Perhaps it’s not true in that our genetics, epi genetics and physical environments are simply specifics of a simpler abstraction of fundamental “starting conditions.” There’s no need to appeal to a distinctly human quality of “human” thinking.

The discussion of verbal behavior/language is useful. Consider the “sms phenomenon”. We’re fully capable of evolving completely rule breaking verbal structures without destroying human communication. txting has structure but it’s hardly ordinary sequential grammatical.

The discussion in chapter 12 is useful as well. Purpose gets in the way of human behavioral analysis. We want to personify so badly, but there’s nothing in the evidence suggesting value in doing so.

Read Full Post »