Sunday, January 29, 2006
Nextumi hires new CEO
I am chief scientist for a web personalization, introduction, and recommendation firm called Nextumi (see earlier post here). The company has hired a new CEO (see here).
A simple GA in Labview
Genetic algorithms have been written in almost every imaginable language, but this blog post at robsyvertsen.org describes a simple GA written in Labview, a graphical programming language mainly used for real-time instrumentation control and data collection. The code is downloadable here for those interested.
Thursday, January 26, 2006
Stanford classes on iTunes
I just saw on Slashdot that Forbes is running an article on Stanford delivering for free lectures on iTunes. Interesting move now that podcasting is changing the way that people interact with media contents.
Seurat by GA
There's a post over at citym.org (here) that uses genetic algorithms to evolve a pointilist rendering of a photograph
Wednesday, January 25, 2006
Presence reviewed
I review Senge's new book Presence over at TEE here.
Congratulations to Martin Pelikan
I just learned that Martin Pelikan has received an NSF Career award for his work on hBOA. The NSF Career program is described here. Martin received his PhD in the lab in 2002 and is author of the book Hierarchical Bayesian Optimization Algorithm: Toward a New Generation of Evolutionary Algorithms. Join me in congratulating Martin on receiving this award.
Tuesday, January 24, 2006
Lunch conversations
Today Moshe Looks is visiting the lab. We are having nice idea exchanges about genetic programming, BOA for continuous domains, competent genetic algorithms, human-computing interaction, and more. But I do not want to blog about them. Actually, over the lunch at Mandarin Wook, a movie sneaked into the conversation: Forbidden Planet (1959). No, no computer, or psychology involved, just a simple issue: data and knowledge storage. In a digital era where we are storing more and more information on electronic devices, a black out on such computer-mediated knowledge access will turn all that stored knowledge into something useless. No human been would be able to access to it. Interesting topic for a casual lunch. Oh, I forgot to mention, Kumara does not like the movie (interesting since Leslie Nielsen is in it).
Friday, January 20, 2006
No rules allowed
Apartment theory is and interesting blog about interior design. No, no GAs there yet, but I just ran into and interesting quote from Tomas Edison: There ain't no rules around here! We're trying to accomplish something. That got me thinking again about structure and its enforcement in creative problem solving. From the work done in the DISCUS project, we started with a free-form approach. That was tough, users tended to get lost. Then we ruled a step-by-step approach. Users felt caged when using such a version. So, as Edison said, there are no rules around here. DISCUS eventually evolved into a middle ground, providing free-form and step-by-step approaches, leaving the freedom to the user for choosing what he/she feels comfortable with. Eventually, DISCUS tools are to support for human creativity and innovation, not to replace it. Yes, this is one of our DISCUS there-ain’t-no-rules-here lessons. Thank you Mr. Edison.
Wednesday, January 18, 2006
GECCO-2006 submissions deadline extended to Feb 1
The submissions deadline for the ACM 2006 Genetic and Evolutionary Computation Conference to be held July 8-12, 2006 (Sat-Wed) in Seattle, WA has been extended to February 1. For further information, consult the GECCO website here.
Sunday, January 08, 2006
Automating invention on quantum computers and GP
Automating invention has a post (here) on quantum computers and its implications for automated invention using GP.
Novamente uses BOA for AI machine
According to Notes on Minds, a company called Novamente uses a Bayesian optimization algorithm (see related post here) as part of its larger artificial intelligence architecture:
The basic idea is to dig for patterns using a bayesian optimization algorithm (a heuristically improved genetic algorithm). Patterns are cloned, mutated, and tested. Good patterns are kept and associated with weights. These weights control how much "attention" each pattern later receives.Novamente was founded by an interesting fellow I might a couple years ago named Ben Goertzel. Ben has a roving and interesting mind, and although his ambitions are large, I wouldn't bet against his efforts.
Later, patterns are connected into a hypergraph, which is a graph, but with special edges that may point to more than two vertices, and be of various types. One edge type indicates that a vertex A is a special case of vertex B; another says that several vertices are similar.
Thursday, January 05, 2006
Values, mission, and goals at TEE
The Entrepreneurial Engineer has a post on making lists of values, missions, and goals for the new year here.
Wednesday, January 04, 2006
Fogel book goes into third edition
See press release here.
Soft computing center started
A soft-computing center, The Centre for Soft Computing Research, has been started at the India Statistical Institute, Kolkata:
The center grows out of very early work in fuzzy systems and logic.
Basic research will be undertaken in areas such as pattern recognition, image processing, computer vision, neural networks, genetic algorithms, data mining, hybrid techniques, video image processing, etc.
The center grows out of very early work in fuzzy systems and logic.
Consumer credit assessment using GAs
Genetic algorithms are used to assess consumer credit worthiness in this talk.
GAs avoid sewer overflows
You heard it here first.
GECCO 2006 submissions deadline approaches
The submissions deadline for the 2006 Genetic and Evolutionary Computation Conference (GECCO-2006) is January 18, 2006. Submission information is posted here. GECCO is sponsored by ACM's newest special interest group, SIGEVO, the special interest group for genetic and evolutionary computation. GECCO 2006 will be held July 8-12, 2006 (Saturday-Wednesday) in Seattle, Washington. Mark your calendars for what promises to be the best genetic and evolutionary computation conference ever.
DISCUS in Japanese
Blogger kenichihorie writes a short post on DISCUS in Japanese here. The post talks about Kazu Inaba's visit at IlliGAL to work on DISCUS.
A Japanese version of DISCUS has been prepared and we hope to use it with in pilot tests with Japanese consumers shortly.
A Japanese version of DISCUS has been prepared and we hope to use it with in pilot tests with Japanese consumers shortly.
Tuesday, January 03, 2006
Don't worry, be happy
Monday, January 02, 2006
Happy 2006
I apologize for no blogging over the holidays. Took mother penguin and the two child penguins to Orlando for four nights to get some sun and fresh air. Went to SeaWorld, sat by the pool, and walked on the ocean. Last trip to Orlando was in 1999. I wonder what Walt Disney would think of what his vision has become? I'm sitting in Espresso Royale (on Goodwin) thinking about the new year, making goals, and getting ready for a big year. I hope your year is laden with ideas and accomplishment.