View unanswered posts | View active topics It is currently Sun Feb 18, 2018 5:15 pm




Reply to topic  [ 1 post ] 
 Intelligence 101 + Free Intelligence Detection Lab 
Author Message

Joined: Fri Sep 22, 2006 12:04 am
Posts: 1456
Location: Massachusetts
Post Intelligence 101 + Free Intelligence Detection Lab
MOLECULAR INTELLIGENCE

Molecular Intelligence is a little known field of science to analyze molecular systems that display intelligence.

With it we find out what makes zooplankton smart. Science teachers would probably love a good answer to one like "Where is a paramecium's brain?". It's very useful science, that is still being pioneered.

At the end of this presentation are references to MI use in academia. In the following tutorial part we will learn how to generate and detect intelligence, using software.


STATE CHANGE

Anything that changes the state of a atom or molecule, makes it somehow different than it was before, is a "state change". Losing or gaining an electron is a state change.

If you have no sound coming out of a loud stereo, then you hear the transistor noise, a hiss. That's from a microscopic bit of semiconductor mineral in an integrated circuit where a flood gate is holding back all the electrical power the stereo has. The hiss are its atoms losing or gaining an electron, changing states, by heat or other force making random state changes that in a perfect system would be 100% efficient so not have a hiss in the music.

A state change is like throwing a switch that causes something to happen. Another electron being knocked out of orbit amplifies into another pop in the speakers.


INTELLIGENCE DETECTION

The pop could have been in the music, say a song about transistor noise. So to fully understand what is random and what is not, you must be able to detect the intelligence. In our case that's easy, would know that a song about transistor noise hiss, has hiss in it on purpose. So the randomness did not go way up for some unexplainable reason, the intelligence did it on purpose.

The randomness part of a radio signal is the hiss, the noise. The nonrandom intelligence is the music, product of human intelligence. When speech that is there is understood, can see it can even predict the weather which can cause humans to flee their habitat in anticipation of a hurricane, and more. An outside observer like an alien picking up a stray radio signal from Earth would see the tell tale signs of nonrandom behavior, intelligence, in it.


HEAT MOTION

In a cell, heat shakes up the molecules so they must overcome the forces pulling them apart. When they do the shaking helps move them to a new place. It's powering like in one of those vibrating tables that make figures move around the table without needing a motor to make them move. When the right molecules find each other they hold together, otherwise they just wander aimlessly until they bump into what they are supposed to, where they will stay as part of the intelligence system.

This motion from heat shaking is one of the things that powers molecules. A state change in physical 3D space. Very important to think of heat as a mover. If you have the simulation working correctly, then cranking the heat to boiling would usually cook it, molecules all rearranged so they can no longer move around like they did, flatline intelligence now detected.


INTELLIGENCE DETECTOR

Notice the graph at the bottom of the intelligence generator program.

Image

And this one with neurons hooked up in a ring so it senses what's out of its field of vision. Learns how to turn in the right direction in response to chase. You can grab the only feeder with your mouse to see how it gains that over time.

Image

http://www.kcfs.org/cgi-bin//ultimatebb ... =001464;p=

The graphs show the success rate and confidence of the intelligence. In this one you can see how well it keeps itself fed, the red line

No intelligence at all would have the graph showing a battery level that flatlines at 0, which in essence, is dead. But notice the red line here. See how it very quickly learns to find food so it doesn't go to zero full, starved, like it would stuck in a circle that never bumps into food.

The way the lines in the graph look, describes the intelligence. Overconfidence is an all of a sudden making more errors, on purpose. Then has a period of lowered confidence but then it climbs to another possible episode of overconfidence.

There is no special memory that has location of objects stored in it, but it still knows where they are, because sensing direction adds another level of intuition. It doesn't need to see what is out of its field of view to know which way it is.

When typing we send the letters to our fingers. They hit the right keys on their own without us thinking about them. Motor actions, that coordinate muscle movement, is a simple way to "see" things like keys on a keyboard. Where everything is, is stored as movements to get there.


THE DOWNLOADABLE LAB

All files are here:

http://members.aol.com/garygaulin/intel ... ator32.zip

Includes source code, written in Visual Basic. Is so simple a loop to make it sense/react that once you know the trick rewriting in any language would be easy. Visual basic is coded like we write sentences, so it's an easy one to follow for those who have no programming experience.

You can now gain experience recognizing intelligence in a system. The contents of its memory can be shown in text that you can copy somewhere to examine. Like this, after a few seconds of awareness:

Code:
Main Memory of Intelligence Generator
10/8/2007  3:47:41 AM
-----------------------------------------------
           ADDRESS                  DATA
-----------------------------------------------
       An  Tf  Fd  St   RF  LF      RF  LF
Addr     Sf  Sp  Fu   RR  LR      RR  LR   Cf
00000  0 0 0 0 0 0 0  0 0 0 0     0 0 0 0  1
00512  0 1 0 0 0 0 0  0 0 0 0     0 1 1 1  3
00515  0 1 0 0 0 0 0  0 0 1 1     1 0 1 1  0
00769  0 1 1 0 0 0 0  0 0 0 1     1 1 0 1  3
00773  0 1 1 0 0 0 0  0 1 0 1     1 1 0 1  0
00775  0 1 1 0 0 0 0  0 1 1 1     0 1 0 1  3
00781  0 1 1 0 0 0 0  1 1 0 1     0 0 1 1  0
00897  0 1 1 1 0 0 0  0 0 0 1     0 0 0 1  3
01024  1 0 0 0 0 0 0  0 0 0 0     0 1 0 1  0
01026  1 0 0 0 0 0 0  0 0 1 0     1 0 1 0  1
01027  1 0 0 0 0 0 0  0 0 1 1     1 1 0 0  0
01032  1 0 0 0 0 0 0  1 0 0 0     1 0 1 0  0
01036  1 0 0 0 0 0 0  1 1 0 0     0 0 1 0  0
01037  1 0 0 0 0 0 0  1 1 0 1     0 0 0 1  0
01038  1 0 0 0 0 0 0  1 1 1 0     0 1 0 1  0
01285  1 0 1 0 0 0 0  0 1 0 1     0 0 1 1  0
01344  1 0 1 0 1 0 0  0 0 0 0     1 0 1 0  3
04097  4 0 0 0 0 0 0  0 0 0 1     1 0 1 0  0
04098  4 0 0 0 0 0 0  0 0 1 0     0 1 0 0  0
04100  4 0 0 0 0 0 0  0 1 0 0     0 0 0 1  0
04106  4 0 0 0 0 0 0  1 0 1 0     1 1 0 0  0
04107  4 0 0 0 0 0 0  1 0 1 1     0 0 0 1  0
04108  4 0 0 0 0 0 0  1 1 0 0     1 1 1 1  0
04110  4 0 0 0 0 0 0  1 1 1 0     1 0 1 1  3
04111  4 0 0 0 0 0 0  1 1 1 1     1 1 1 0  0
05120  5 0 0 0 0 0 0  0 0 0 0     1 1 0 1  0
05121  5 0 0 0 0 0 0  0 0 0 1     1 1 0 0  0
05123  5 0 0 0 0 0 0  0 0 1 1     0 1 0 1  0
05124  5 0 0 0 0 0 0  0 1 0 0     1 0 1 0  0
05127  5 0 0 0 0 0 0  0 1 1 1     0 0 0 1  0
05128  5 0 0 0 0 0 0  1 0 0 0     0 0 0 1  3
05132  5 0 0 0 0 0 0  1 1 0 0     0 0 1 1  0
05133  5 0 0 0 0 0 0  1 1 0 1     1 0 0 0  0
05134  5 0 0 0 0 0 0  1 1 1 0     0 1 1 1  0
05381  5 0 1 0 0 0 0  0 1 0 1     1 0 0 0  0
-----------------------------------------------
Conf 3     7
Conf 2 <3  0
Conf 1 <2  2
Conf   <1  0
Conf 0     26
-----------------------------------------------


The "Address" is a number you get when you turn the sensory input into a number that makes it unique. How you combine the sensory input changes the number you end up with, but it doesn't matter. Only has to address a unique location, to store current environmental sensory information.

The "Data" droves it's motors, muscles, molecules, atoms, etc.. In this simulation, the critter can one side or the other go in either direction or stop if there is a memory of it being a good thing to do, like when hungry and at a feeder.

The confidence (Cf) is a number from 0 to 3 that describes how reliable the action data is. A new unique sensory experience is 0 which is the cleared state of a memory. Lifetime starts with everything in memory zero. No memory of anything exists at that point.

Cf is incremented when the actions stored in Data led to success. In this model reversing after hitting one of four sides of the square that contains it, reinforces it's confidence in that action.

The motor controls like Right Motor Forward (RF) or Right Motor Reverse (RR) are in both Address and Data. The address is sensory feedback information, back from the motor circuit. To for example know if the motor is turning. Could have it switched on but when stuck against a wall it's stalled, not turning. In a new situation it has never seen before it takes a guess what to do, eventually finding out that setting motor in reverse works, so it reinforces that action in a certain situation by increasing confidence number.

There is no motor Off register. That's what happens if both RMF and RMR are 0 or 1. Need to be 0 and the other a 1 for a motor to start turning.

A single celled animal could be simulated this way. Motors that makes them move are in flagella, microtubules, heat motion, polar force, and such.


EVOLUTION CONNECTION

We can parallel the evolution mechanism with this software. Both have a random component that takes guesses and a memory that stores responses that work.

In the program the critter takes random guesses that are stored in arrays. In evolution there is random mutation over time. Evolution does not need an array for it to only remember responses that work, only those that survive to reproduce, remain through time.


ACADEMIC EVIDENCE FOR MI

Quote:
Description
The phenomenal rate of progress in the fields of molecular and cellular biology is rapidly creating new opportunities for molecular bioengineering to advance medicine and biology. Fundamental advances in our understanding of how biomolecules control healthy physiology, how cells communicate, how molecular machines are coupled to force and signal transduction, how the body protects itself against the initiation of cancer, etc. are opening many new avenues for developing medical technologies. The expansion of the Molecular Bioengineering Program to include Nanotechnology arises from the natural synergy between molecular bioengineering research and the new infusion of nanotechnology research, with several Bioengineering faculty research programs already straddling both areas. The Molecular and Nanotechnology Thrust brings some of the broader nanoengineering advances found in chemistry, physics, and biology into the bioengineering field for the advancement of medicine. This Thrust also contains strong programs in muscle contraction and engineering of biopolymer gels that complement those of the molecular bioengineering faculty.

Areas of Research in Molecular Bioengineering & Nanotechnology
The Molecular and Nanotechnology Thrust Area encompasses a broad variety of research areas. Some are directed toward more fundamental research that is tied to identifying and better understanding the barriers to medical advancement, while others are directed toward translational research and development of better medical technologies such as drug delivery and diagnostics. A unifying theme is bioinspiration and biomimicry, where nature's mechanisms for molecular interactions and medicine are both studied and tied to technology development.

Smart Delivery (Hoffman, Pun, Ratner, Stayton).
The next generation of delivery systems must be quite "smart," in the sense that they must often be able to target specific cells, transverse biological transport barriers, and control release levels in response to physiological or external signals. A principal property of biological molecules and systems is their ability to change properties in response to stimuli. For example, the high gene transfection efficiency of viruses arises from their unique fusogenic proteins that sense the lowered pH of endosomes and respond by becoming membrane active to enhance transport of DNA across the membrane barrier. It is this ability to sense and respond that provides control over a wide variety of cellular processes, and this type of molecular intelligence also represents an important design goal for many drug delivery strategies. Stayton and Hoffman are developing nanoengineered polymeric materials that are bioinspired by viral/toxin strategies, and which have built-in stimuli-responsiveness to control and enhance intracellular trafficking of biomolecules such as proteins, peptides, RNA, and DNA and their targeting to specific compartments. Another example is from Ratner and Crum, where molecular self-assembly strategies are being developed to create molecularly engineered coatings that sense ultrasound signals and release controlled doses of drugs from underlying implants. The Pun group focuses on advancing macromolecule drug delivery technology by developing materials that use biomimetic approaches to overcome transport limitations in tissues and within cells.

http://depts.washington.edu/bioe/resear ... /nano.html


Quote:
Course 4190.626: Molecular Intelligence (Knowledge Representation and Reasoning) School of Computer Science and Engineering, Seoul National University

http://bi.snu.ac.kr/Courses/g-ai02/g-ai02.html

Quote:
Publications on Molecular Intelligence

Probabilistic Library Model
Molecular learning of wDNF formulae, Byoung-Tak Zhang and Ha-Young Jang, Lecture Notes in Computer Science, , vol. 3892, pp. 427-437, 2006. [PDF]
A Bayesian Algorithm for In Vitro Molecular Evolution of Pattern Classifiers, Byoung-Tak Zhang and Ha-Young Jang, Lecture Notes in Computer Science , vol. 3384, pp. 458-467, 2005. [PDF]
Molecular Programming
Temperature Gradient-Based DNA Computing for Graph Problems with Weighted Edges, Lee, J. Y., Shin, S.-Y., Augh, S. J., Park, T. H., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2568, pp. 73-84, 2003. [PDF]

Molecular Immunocomputing with Application to Alphabetical Pattern Recognition Mimics the Characterization of ABO Blood Type, Kim, S. D., Shin, K.-R., and Zhang, B.-T., Proceedings of the 2003 Congress on Evolutionary Computation, vol. 4, pp. 2549-2556, 2003.
Molecular Algorithms for Efficient and Reliable DNA computing, Zhang, B.-T. and Shin, S.-Y., Proceedings of the Third Annual Genetic Programming Conference (GP-98), Koza, J.R. et al. (Eds.), Morgan Kaufmann, pp. 735-742, 1998.

Molecular Learning
Version Space Learning with DNA Molecules, Lim, H.-W., Yun, J.-E., Jang, H.-M., Chai, Y.-G., Yoo, S.-I., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2568, pp. 143-155, 2003. [PDF]
Molecular Inference
RCA-Based Detection Methods for Resolution Refutation, Lee, I.-H., Park, J. Y., Chai, Y.-G., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2943, pp. 32-36, 2004.
Oligonucleotide-Based Theorem Proving by Cross-Linking Gold Nanoparticle Assembly, Park, J.-Y., Kim, T.-G., Lee, I.-H., Park, J.-G., Zhang, B.-T., and Chai, Y.-G., Preliminary Proceedings of the Tenth International Meeting on DNA Computing, p.446, 2004
DNA Implementation of Theorem Proving with Resolution Refutation in Propositional Logic, Lee, I.-H., Park, J.-Y., Jang, H.-M., Chai, Y.-G., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2568, pp. 156-167, 2003. [PDF]

Molecular Search
Solving Travelling Salesman Problems Using Molecular Programming, Shin, S.-Y., Zhang, B.-T., and Jun, S.-S., Proceedings of the 1999 Congress on Evolutionary Computation (CEC99), vol. 2, pp. 994-1000, 1999. [PDF]

Design Tools (NACST)
NACST/Seq: A Sequence Design System with Multiobjective Optimization, Kim, D., Shin, S.-Y., Lee, I.-H., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2568, pp. 242-251, 2003. [PDF]
DNA Sequence Optimization Using Constrained Multi-Objective Evolutionary Algorithm, Lee, I.-H., Shin, S.-Y., and Zhang, B.-T., Proceedings of the 2003 Congress on Evolutionary Computation, vol. 4, pp. 2270-2276, 2003.
DNA Computing Complexity Analysis Using DNA/DNA Hybridization Kinetics, Shin, S.-Y., Lee, E. J., Park, T. H., and Zhang, B.-T., Preliminary Proceedings of the Ninth International Meeting on DNA Based Computers, p. 207, 2003. [PS]
Evolutionary Sequence Generation for Reliable DNA Computing, Shin, S.-Y., Kim, D.-M., Lee, I.-H., and Zhang, B.-T., Proceedings of the 2002 Congress on Evolutionary Computation (CEC2002), vol. 1, pp. 79-84, 2002. [PDF]
Code Optimization for DNA Computing of Maximal Cliques, Zhang, B.-T. and Shin, S.-Y., Advances in Soft Computing: Engineering Design and Manufacturing, Springer-Verlag, 1998. pp. 735-742, Morgan Kaufmann, 1998

DNA Computing Chips
Ligation Module for In-vitro Selection in DNA Computing, D. van Noort, I. H. Lee, L. F. Landweber, and B. T. Zhang, SPIE International Symposium on Smart Materials, Nano-, and Micro-Smart Systems, 2004. [PDF]
PDMS Valves in DNA Computers, D. van Noort and B.-T. Zhang, SPIE International Symposium on Smart Materials, Nano-, and Micro-Smart Systems, 2004. [PDF]
A Lab-on-a-Chip Module for Bead Separation in DNA-Based Concept Learning, Lim, H.-W., Jang, H.-M., Ha, S.-M., Chai, Y.-G., Yoo, S.-I., and Zhang, B.-T., Lecture Notes in Computer Science, vol. 2943, pp. 1-10, 2004.

Molecular Evolutionary Computing Lab-on-a-Chip: A First Design and Simulation, Zhang, B.-T., Jang, H.-Y., Cho, D.-Y., and Lee, I.-H., Proceedings of the 10th International Conference on Neural Information Processing, pp. 516-519, 2003.
Effectiveness of Denaturation Temperature Gradient-Polymerase Chain Reaction for Biased DNA Algorithms, Lee, J. Y., Zhang, B.-T., and Park, T. H., Preliminary Proceedings of the Ninth International Meeting on DNA Based Computers, p. 208, 2003. [PDF]

--------------------------------------------------------------------------------
This page is maintained by Byoung-Hee Kim (Email me)
Last update: February 23, 2007.

http://bi.snu.ac.kr/Publications/pub_mi.html


Mon Oct 08, 2007 5:58 am
Profile
Display posts from previous:  Sort by  
Reply to topic   [ 1 post ] 

Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by STSoftware for PTF.