Meetup on Cross-Entropy method for integration and optimization

Meetup on Cross-Entropy method for integration and optimization

Leibniz bemoaning excess of knowledge in books

Heard in interview with James Gleik on Science Friday

http://www.casualoptimist.com/2011/03/11/something-for-the-weekend-52/

In the 17th century Leibniz bemoaned the “horrible mass of books which keeps on growing,” and in 1729 Alexander Pope warned of “a deluge of authors cover[ing] the land,” as James Gleick describes in his new book, The Information.

Definition of a game

Heard on Science Friday (or was it Sci Am?) about 9 months ago: unnecessary obstacles

Oger: another machine learning toolbox

R: simply start over and build something better

One interesting comment:

Avram Says:
September 13, 2010 at 10:21 pm

I have been (and remain) a long-time R user and fan. Lately, however, I find myself learning/using Clojure (clojure.org) with Incanter (http://incanter.org/).

The main reasons for this are (1) memory issues for big data using R, (2) the paradigm of a language deeply rooted in functional programming, (3) concurrency, (4) Java-interoperability as Clojure runs on the JVM, and (5) immediate use of the vast number of incredible Java libraries available (e.g. hadoop, mahout, weka, etc…), to name a few.

I would be very curious as to how Ross and others view Clojure in the space of the future of statistical programming.

Many thanks.

Reply

Development 2.0: Give me less information and more data!

about relative value of raw data and distilled information

Book: Applied Computational Intelligence: How to Create Value

2.2 Applied Computational Intelligence: How to Create Value
Posted by: Arthur Kordon (akordon@dow.com)
Date submitted: Nov. 24th, 2009

Content:
Applied Computational Intelligence: How to Create Value
Arthur Kordon
Springer Verlag, 481 pp., Hardcover, December 2009
ISBN: 978-3-540-69910-1
http://www.springer.com/978-3-540-69910-1

DESCRIPTION
The flow of academic ideas in the area of computational intelligence is
impacting industrial practice at considerable speed. Practitioners face the
challenge of tracking, understanding and applying the latest techniques,
which often prove their value even before the underlying theories are fully
understood. This book offers realistic guidelines on creating value from the
application of computational intelligence methods.

In Part I, the author offers simple explanations of the key computational
intelligence technologies: fuzzy logic, neural networks, support vector
machines, evolutionary computation, swarm intelligence, and intelligent
agents. In Part II, he defines the typical business environment and analyzes
the competitive advantages these techniques offer. In Part III, he
introduces a methodology for effective real-world application of
computational intelligence while minimizing development cost, and he
outlines the critical, underestimated technology marketing efforts required.
The methodology can improve the existing capabilities of Six Sigma, one of
the most popular work processes in industry. Finally, in Part IV the author
looks to technologies still in the research domain, such as perception-based
computing, artificial immune systems, and systems with evolved structure,
and he examines the future for computational intelligence applications while
taking into account projected industrial needs.

The author adopts a light tone in the book, visualizes many of the
techniques and ideas, and supports the text with notes from successful
implementations. The book is ideal for engineers implementing these
techniques in the real world, managers charged with creating value and
reducing costs in the related industries, and scientists in computational
intelligence looking towards the application of their research.

Table of Contents:
Part I: Computational Intelligence for the Masses
Chapter 1: Computational vs. Artificial Intelligence.
Chapter 2: A Roadmap Through the Computational Intelligence Maze.
Chapter 3: Fuzzy Systems: Lets Get Fuzzy.
Chapter 4: Machine Learning: The Ghost in the Learning Machine.
Chapter 5. Evolutionary Computation: The Profitable Gene.
Chapter 6: Swarm Intelligence: The Benefits of the Swarms.
Chapter 7: Intelligent Agents: The Computer Intelligence Agency (CIA).
Part II: Computational Intelligence Creates Value
Chapter 8: Why We Need Intelligent Solutions?.
Chapter 9: Competitive Advantages of Computational Intelligence.
Chapter 10: Issues in Applying Computational Intelligence.
Part III: Computational Intelligence Application Strategy
Chapter 11: Integrate & Conquer.
Chapter 12: How to Apply Computational Intelligence?.
Chapter 13: Computational Intelligence Marketing.
Chapter 14: Examples of Computational Intelligence Industrial Applications.
Part IV: The Future of Applied Computational Intelligence
Chapter 15: Future Directions of Applied Computational Intelligence
Glossary
Index

Arthur K. Kordon is a Data Mining and Modeling Leader in the Data Mining and
Modeling Capability of The Dow Chemical Company. He is an internationally
recognized expert in applying emerging technologies in industry, and has
given talks and chaired panels on the topic at the major computational
intelligence conferences such as WCCI and GECCO. He has successfully
introduced several novel technologies for improved manufacturing and new
product design in the chemical industry, and his research interests include
application issues of computational intelligence, robust empirical modeling,
intelligent process monitoring and control, and data mining.
=========================================================

Hash Kernels for Structured Data

Qinfeng Shi, James Petterson, Gideon Dror, John Langford, Alex Smola, and SVN Vishwanathan, Hash Kernels for Structured Data, AISTAT 2009 and JMLR 2009.

Representation and structure in langauge processing

Reith lectures

these are available as BBC radio podcasts in iTunes

Lego First competition

Graph Grammars

Graph Grammars Workshops
Handbook of Graph Grammars and Computing by Graph Transformation (note there are 3 volumes)

special issue: Perspectives on Social Cognition

Cognitive Systems Research
Volume 9, Issues 1-2, Pages 1-160 (March 2008)
special issue: Perspectives on Social Cognition

One author has papers at:
http://www.indiana.edu/~smithlab/sitcogarticles.shtml

Religion is a product of evolution, software suggests

(New Scientist)
http://jasss.soc.surrey.ac.uk/11/2/2.html

Powerset (natural language search)

http://searchengineland.com/080512-000100.php
http://searchengineland.com/080103-084033.php
http://www.techcrunch.com/2008/05/11/powerset-launches-showcase-for-user-search-experience/

On income inequality

and http://streetlightblog.blogspot.com/2007/05/income-inequality-international.html

On search in pubmed

Graph rewriting system resources

Have: http://citeseer.ist.psu.edu/746463.html Visual Programming with Graph Rewriting Systems (1995)
A. Schürr, A. Winter, A. Zündorf

Needs subscription to get this paper: http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel4/5725/15315/00706131.pdf?temp=x
A graph rewriting programming language for graph drawing
Rodgers, P.J.
Visual Languages, 1998. Proceedings. 1998 IEEE Symposium on
Volume , Issue , 1-4 Sep 1998 Page(s):32 - 39
Digital Object Identifier 10.1109/VL.1998.706131
Summary:The paper describes Grrr, a prototype visual graph drawing tool.

Book on Amazon: http://www.amazon.com/Efficient-Rewriting-Implementation-Lecture-Computer/dp/3540600558
Efficient Graph Rewriting and Its Implementation (Lecture Notes in Computer Science) (Paperback)
by Heiko Dörr (Author)

Graph Rewriting Semantics for Functional Programming Language

Reasoning with mental models

Mental models: a gentle guide for outsiders
by
P.N. Johnson-Laird, Vittorio Girotto, and Paolo Legrenzi
http://www.si.umich.edu/ICOS/gentleintro.html

Neural-Net Implementation of Complex Symbol-Processing in a Mental Model Approach to Syllogistic Reasoning. John A. Barnden* ... Conposit is a neural net system that is aimed at high ...
http://dli.iiit.ac.in/ijcai/IJCAI-89-VOL1/PDF/091.pdf

Evolution of Language from Theory of Mind or Coevolution of Language and Theory of Mind? Anne Reboul

An Evolutionary Post Production System

(by Eric Baum)

and lots more interesting stuff @ http://www.whatisthought.com/eric.html

Jackendoff recent papers

Links for evolution of cooperation

http://en.wikipedia.org/wiki/Evolutionary_game_theory
http://www.dmoz.org/Science/Social_Sciences/Economics/Game_Theory/Evolutionary_Game_Theory/
http://users.tpg.com.au/users/jes999/5.htm

Brian Arthur on Technology Creating Technology

Wednesday, November 14, 2007
7:30 PM at the James A Little Theater

W. Brian Arthur SFI External Professor and Visiting Researcher, Intelligent Research Labs, PARC.

with Discussant J. Doyne Farmer, SFI Professor


Technology Creating Technology

Technology — the collection of devices and methods available to us as humans — grows over historical time by a self-reinforcing mechanism. Novel technologies are created out of building blocks that are themselves technologies, and they go on to become potential building blocks for the construction of further new technologies. I n this sense, technology creates itself out of itself. Arthur will explore these ideas by looking at the evolution of technology both in human history and in an artificial computer world. In both cases, technology builds by bootstrapping itself from few building-block elements to many, and from simple elements to complicated ones. Arthur will also discuss links with biological evolution.

Peter Naur on "computing vs human thinking"

In CACM, Jan 2007, V50N1, p85-93. Has some interesting comments about the nature of language & cognition, & refs to William James & Otto Jespersen

Software sandboxes

"Safe" module (ch 23.3 in "Programming Perl, 3rd ed) (sandbox)
Ruby Sandbox module http://code.whytheluckystiff.net/sandbox/
Steel Bank Common Lisp http://www.sbcl.org/
http://www.cl-user.net/asp/root-dir (apparently, Common Lisp has a sandbox, but I couldn't find it...)

Evolution of morality

Some interesting ideas here about the evolution of morality (& language...), e.g.:

Furthermore, Damasio and Bargh both found, as Michael Gazzaniga had years before, that people couldn't stop themselves from making up post-hoc explanations for whatever it was they had just done for unconscious reasons. Combine these developments and suddenly Kohlbergian moral psychology seemed to be studying the wagging tail, rather than the dog. If the building blocks of morality were shaped by natural selection long before language arose, and if those evolved structures work largely by giving us feelings that shape our behavior automatically, then why should we be focusing on the verbal reasons that people give to explain their judgments in hypothetical moral dilemmas?

...

Moral thinking is for social doing. This is a play on William James' pragmatist dictum that thinking is for doing, updated by newer work on Machiavellian intelligence. The basic idea is that we did not evolve language and reasoning because they helped us to find truth; we evolved these skills because they were useful to their bearers, and among their greatest benefits were reputation management and manipulation.

2007 Info retrieval book

Compressed sensing, sparse neuronal activation

Good summary of some issues re sparse neuronal activation.

The Discover Interview: Marvin Minsky

Is there other work in neuroscience or AI that interests you?

Very little. There are 20,000 or 30,000 people working on neuronetworks, and there are 40,000 or 50,000 people working on statistical predictors. There are several thousand people trying to get logical systems to do commonsense thinking, but as far as I know, almost none of them can do much reasoning by analogy. This is important because the way people solve problems is first by having an enormous amount of commonsense knowledge, like maybe 50 million little anecdotes or entries, and then having some unknown system for finding among those 50 million old stories the 5 or 10 that seem most relevant to the situation. This is reasoning by analogy. I know of only three or four people looking at this, but they're not well-known because they don't make grandiose claims of looking for a theory of everything.

PS3 architecture

http://www-128.ibm.com/developerworks/power/library/pa-cbea.html

Feynman lectures online

Review of Jerry Feldman's new language book

Jerome Feldman (2006), From Molecule to Metaphor: A Neural Theory of Language (Cambridge: MIT Press: 0262062534)

password hash

Firefox plugin: https://addons.mozilla.org/firefox/1033/

Mondoglobo

podcasts

Animated series puts all files under CC license

http://www.boingboing.net/2006/07/26/animated_series_puts.html
Animated series puts all files under CC license
John sez, "We just posted a story over on Drawn! about the Canadian animated series Odd Job Jack, which is releasing the master Flash files and bitmaps of every piece of art used in this season of the show: every character, prop, and background from every episode plus tutorials and other support material, all free to hack, use, remix under a share-friendly license."

Why: We love animation and we just know you do too. We're proud of Odd Job Jack and we've put lots of work into our show. Our art deserves to live beyond broadcast and who better to give a free gift to than the entire planet?

When: Every Monday during our 13 episode broadcast we will release a new set of files. First episode air July 22nd. The torrent will be available the following Monday.

Put the moon into orbit -- neat svg simulation

Ning web application hosting

Researchers Teach Robots To Evolve Their Own Language

Researchers Teach Robots To Evolve Their Own Language


Researchers from Sony and elsewhere are working to give robots linguistic and cognitive skills that grow and develop spontaneously over time, without communication rules from humans.


By Paul McDougall
InformationWeek

Jul 10, 2006 12:00 AM

Sony killed off its Aibo line of robotic dogs this year, but scientists with the consumer electronics giant are spinning off a litter with new brains. The research could ultimately produce far more advanced forms of artificial intelligence.

Researchers at Sony's computer science labs in France are working with the European Commission's Emerging Technologies Initiative and the Institute of Cognitive Science and Technology in Italy under a program called EC (embedded and communicating) Agents. Their mission: Give robots linguistic and cognitive skills that develop on their own over time, without communication rules from humans.

In one application of the technology, the researchers gave a pack of Aibos the ability to develop their own language. So, for instance, one dog can tell another the location of a ball and ask it to go fetch. It can also specify whether the ball is rolling.

Making sounds helps the super-Aibos fulfill a programmed requirement to learn more about what's around them. The canine cyborgs were placed in a room with other objects, some of which would respond to sounds. A stuffed elephant and other objects would not respond. The dogs ultimately spent more time "barking" at the responsive objects, and they learned that particular bark patterns elicited certain responses.

The researchers think such interactions are the beginnings of artificial intelligence with an innate language capability--one they hope to develop further. "The result is machines that evolve and develop by themselves without human intervention," Dr. Stefano Nolfi, the coordinator of the EC Agents project, says in a statement posted on the project's site.

Another participant in the project, Sweden's Viktoria Institute, is using EC agents to let mobile devices, such as MP3 players and cell phones, "talk" to each other. That way, a device belonging to, for example, a person who's part of an online social group would know what songs are played frequently by devices owned by other group members and could automatically push those tunes to the person.

The Swiss Federal Institute of Technology is teaching hordes of small, wheeled robots embedded with EC agents to swarm over areas to help with search and rescue operations. "We've managed to ground AI in reality, in the real world," says Nolfi, "solving one of the crucial problems to creating truly intelligent and cooperative systems."

Global development statistics presented in interactive bubble charts

Also http://www.gapminder.org/

Kinetic sculpture

Kinetic sculpture hanging balance game (flash)

3 quarks daily on Jeff Hawkins' "On Intelligence"

Brad DeLong "Morning Coffee" on International captial flows

Di Filippo's story "Little Worker" as a podcast

Di Filippo's story "Little Worker" as a podcast
Today the Escape Pod podcast included a wonderful short story by Paul Di Filippo, a consistently great science fiction writer. The story is "Little Worker" and it's pure gold Di Filippo, a reprint from his collection of bio-punk stories Ribofunk The reading by Jonathon Sullivan is likewise stellat. Escape Pod features some great fiction, but with this story, they've really gone to a new level. Bravo!

At home, Little Worker could do pretty much as she pleased, as long as she was there should Mister Michael need her. At the office-and in other public places-she had to be more circumspect and diligent. Little Worker was on duty her, in a way that was more intense than behind the electrified fence and active sensors of the estate. (Once, one of the men at the Training School had said: "Little Worker, you are the most diligent companion I've ever trained." The men of the school had been nice, in their stern way. But no one was like Mister Michael.)

Today, however, Little Worker's mind was not on her work.

Things That Go Bump in Physicists' Night

http://www.marginalrevolution.com/marginalrevolution/2005/11/things_that_go_.html

Things That Go Bump in Physicists' Night

Truth be told, physicists are terrified of quantum mechanics. Really. The rules of quantum calculation seem so strange that anyone afraid of losing his or her mind should be scared. (Those who love to lose their minds, on the other hand, adore it.)

Struggling to make the quantum rules square with a reality "out there," many physicist's position is "shut up and calculate." Others have abandoned standard logic, probability, or decision theory for "quantum" versions of these things, or have decided that consciousness must play a fundamental role. (There is even a quantum game theory.)

In eleven days I give my first talk at a physics department, on my conservative research program that tries to have it all: the quantum rules, a reality out there with no special role for consciousness, and keeping standard logic, probability, and decision theory. I'm not quite there yet, and I may be too close to my work to be objective, but I feel I'm very close.

Of course we can't make all the quantum strangeness go away. For example, reality seems to be intrinsically non-local, and it seems to be far larger than we ever imagined. But the universe we are all familiar with now is far larger than our ancestors ever imagined, and even Newton gave up on locality.

Fear not the quantum night - it really will all make sense someday.

Douglas Baird's podcast of "Coase's Journey"

http://agoraphilia.blogspot.com/2006/02/coasean-blues_23.html
The Coasean Blues
posted 8:14 PM by Glen Whitman

The weirdest thing about Douglas Baird's podcast of "Coase's Journey" (aside from the disturbing news that I, along with most of the economics profession, have been pronouncing 'Coase' incorrectly for all these years) is what you'll find out if you drop the MP3 file into iTunes. The lecture's genre classification? Blues.

http://webcast-law.uchicago.edu/2006/winter/baird-cbi.mp3

Helicopter photos of Mexico city

Helicopter shots of Mexico City neighborhoods -- stunning and unreal
A former Mexico City helicopter pilot has posted a stunning gallery of his aerial photos of the city and environs. Most interesting are the houses, which range from dire slums to these incredible, cookie-cutter low-income houses that look like grids of Monopoly houses. There's lots else to love in this gallery, besides. Link
http://homepage.mac.com/helipilot/PhotoAlbum31.html

Mechanical computer kit from 1960s available again

http://www.boingboing.net/2006/02/21/mechanical_computer_.html
http://www.mindsontoys.com/kits.htm?dc1_main.htm

Mechanical computer kit from 1960s available again
A toy company has reintroduced the Digicomp, a mechanical computer kit from the 1960s. This looks like an amazing educational toy. As Retrothing notes:


The Digicomp is a plastic mechanical computer from the 1960s. It offered three bits of tabletop computing, back in an age where corded telephones were considered high-tech. The machine arrived in kit form; your first task was to assemble the jumble of tubes, rods, and elastic bands into something that resembles a Jetson's parking garage. Once complete, it's a fantastic hands-on way to teach Boolean algebra and binary numbers.

Specialized Search Tools for Academic Content

http://blog.searchenginewatch.com/blog/060213-075310

When you're looking for magazine or journal articles, search engines can be helpful, but other specialized search tools are often a better bet—particularly in the academic, scholarly and sci-tech areas. Guest writer Mary Ellen Bates looks at a number of these highly focused services in today's SearchDay article, Finding Articles Online.
Posted by Chris Sherman on Feb. 13, 2006 | Permalink

CD burning

from ResearchBuzz http://www.researchbuzz.org/2006/02/researchbuzz_roundup_021406.shtml
Thanks to Robert Richardson for his pointer to BurnCDCC, which is another freeware utility that lets Windows users burn ISO files to CD or DVD -- just in case you want to make an Ubuntu or Knoppix or whatever disc.

SF story podcasts

http://www.boingboing.net/2006/02/15/corys_human_readable.html

Cory's Human Readable podcast concludes
I've just posted the concluding installment in the podcast of me reading my novella, Human Readable, a story originally published in the Future Washington anthology and now short-listed for the Locus Award and forthcoming in the Science Fiction Book Club's Best Short Novels of 2005 anthology. The story was podcast in seven parts, and you can pick them all up by subscribing to the podcast feed or by hitting these links below: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7

whohooo!

looking at a pantone formula guide solid coated....

From internet explorer -- can I post an image?

Let's see if I can include an image from internet explorer:

Ok, the image button pops up a window saying blogger.com does not host images...

How do I post pictures?
Blogger does not directly provide image hosting, although you are welcome to include any images in your blog if you have them stored online elsewhere. However:

• Windows users can also use the Hello BloggerBot to post pictures directly to their blogs.
  
• Mac users (and anybody else, for that matter!) can use Flickr to photoblog; see this article for details.
  

Once you have an image online, you can include it in either your template or in a specific post by typing in the following tag: <img src="https://lh3.googleusercontent.com/blogger_img_proxy/AEn0k_t2lVGgV7NV_IyRJOYPVjb5pjWM_R-gsUXLImmi88Frv2YUFxdhyRfsLXwNhORfy-Gtr5ijkBx5S5_ANGDASYy-k124bAOd=s0-d">Just replace the URL here with the actual URL of your image.

Trying to put up a posting

I'm just typing away here inside a web browser. There is a button to upload an image but it doesn't seem to work in firefox. Oh well.