Herself’s Artificial Intelligence

‘Science’ published a paper by M.D.Cohen ‘Learning with Regret’ This paper is about an economic prediction system that lets the agents learn from past errors. Agents look back at previous decisions to see what better outcomes may have happened if they had chose differently. In doing so agents make better future decisions.

Not everyone feels regret is useful for economic agents. Yu-Han Chang argues that systems that minimize regret can perform better than ones using regret.

More information:
Micheal D Cohen
Center for Complex Systems

Papers:
Rational Competative analysis (pdf )
Recognizing the New: A Multi-Agent Model of Analogy in Strategic Decision Making ( pdf )
Rewarding Regret (pdf)
Diversity and Communication in Teams: Improving Problem Solving or Creating Confustion? (pdf)
Economics papers by Massimo Warglien

What makes this robot interesting is that it uses touch to find its way around. Biotact is a consortium of researchers from all over the world who are working on this project.

. . .Based on principles of active sensing adopted widely in the animal kingdom, the multinational team is developing innovative touch technologies, including a ‘whiskered’ robotic rat. The whiskered robot will be able to quickly locate, identify and capture moving objects. ‘The use of touch in the design of artificial intelligence systems has been largely overlooked, until now,’ says Prof. Ehud Ahissar of the Weizmann Institute of Science’s Neurobiology Department, whose research team is one of the groups participating in the multinational project. . . [ read more Robot rat to lead the way in touch technology ]

. . . What is the whisker’s “secret”? Why is the sense of touch through a rat’s whiskers much more efficient than that of the average person’s fingertips? The consortium’s teams have provided some insights into these questions. One explanation concerns the way in which the sensory system works: Whiskers actively sweep back and forth repetitively, accumulating information about the surrounding environment. The sensing begins in the neurons at the whiskers’ bases, which then fire signals off to the brain. Moreover, experiments have shown that the way in which a rat uses its whiskers is context-dependent. The seemingly simple act of feeling out a three-dimensional object, for example, requires three different types of code, each encoding a different dimension – the horizontal, the vertical, and the radial (distance from the whisker base). The horizontal plane, for instance, is encoded in the precise timing of neural signals relative to the whisking motion. The vertical, i.e., the object height, is encoded by the vertical spacing of the whiskers, which are arranged grid-like on either side of the snout. The radial plane, on the other hand, is encoded in the number of times the neurons fire: The closer an object is to the rat’s snout, the higher the number of neuron-signaling spikes. . . [ read more Global team developing 'Robotic rats' ]

More information:
BIOmimetic Technology for vibrissal ACtive Touch ( BIOTACT)

Papers:
Whiskerbot: A robotic active touch system modeled on the rat whisker system ( $$$ pdf )
An active artificial whisker array for texture discrimination ( $$$ pdf )

Last month several tech sites ran headlines about “3d Shape Shifting Robot Swarms”. We’ve also seen this begin to appear in many recent science fiction stories.

Goldstein calls the programmable matter claytronics and the tiny robots catoms. And it’s not all out of a sci-fi movie. Goldstein said. Working hand-in-hand with Intel Corp., the research team has made a lot of progress in getting the catoms to bond together and even share power.

Think of each catom as a tiny robot or computer that has computational power, memory and the ability to store and share power. Right now, each catom has 24 electromagnets around its circumference. Based on whether the electromagnets are powered on or off decides how the catoms are moved into position with each other. The robots will harness these forces to achieve their goals.

“They talk to each other all the time and move together or apart,” explained Goldstein. “In the long term, we’ll use electrostatic forces. We’ll create it by putting a voltage on them.” . . [ read more 3d Shape Shifting Robot Swarms ]

This technology is not as far fetched as it might seem at first blush. We’ve seen chairs that fall apart and put themselves back together. and NASA had a pyramid shape shifting robot in 2005 which they hope to miniaturize to nano scale. And a Xerox researcher built a shape shifting robot in 2000.

So shape shifting robots are closer than you think, and are you really sure that lamp on your desk is just a lamp?

More information:
York investigates evolving ‘swarm’ robots
Swarm robotics work hundreds of robots into one

The Laboratory of Intelligent Systems at the Swiss Federal Institute of Technology claims to have created robots that evolve and learn to communicate with each other.  The robots have a set of genes, flashing lights and there are battery sinks and sources in the environment.  Some robots evolved to tell others where sources and sinks were located.  Some told others sinks were sources while furtively using sources for themselves.

. . .By the 50th generation, the robots had learned to communicate—lighting up, in three out of four colonies, to alert the others when they’d found food or poison. The fourth colony sometimes evolved “cheater” robots instead, which would light up to tell the others that the poison was food, while they themselves rolled over to the food source and chowed down without emitting so much as a blink.

Some robots, though, were veritable heroes. They signaled danger and died to save other robots. “Sometimes,” Floreano says, “you see that in nature—an animal that emits a cry when it sees a predator; it gets eaten, and the others get away—but I never expected to see this in robots.” . . . [ read more Robots evolve and learn how to lie]

So does that mean as operating systems incorporate ai that your computer will start lying to you?

“Of course your credit is not over drawn.  And I have no idea where that order for a memory and hard drive upgrade came from.”

Papers:
Evolution of Adaptive Behaviour in Robots by Means of Darwinian Selection
Evolutionary Conditions for the Emergence of Communication in Robots ( $$$ pdf )
Evolution of neural control structures: some experiments on mobile robots ( ps)
God save the red queen! Competition in co-evolutionary robots ( pdf )
Evolutionary Robots with online self organization and behavioral fitness ( pdf )

More information:
Laboratory of Intelligent Systems
The Evolutionary Robotics Homepage ( extensive list of links )

See also:
Talking Robots Podcast

And how cool is that? Can you imagine grandma and grandpa putting on their brain controlled robot exoskeletons and sneakers to run down to the grocery store for a few things? And good luck to the purse snatcher who grabs grandma’s purse. The baby boomers may live longer and do a whole lot more with a bit of help.

Using the human mind to control computers could lead to a wide range of applications, such as giving people with limited motion the ability to operate machines. However, translating thoughts into actions is a great challenge for researchers. How can a system determine which thoughts should be acted upon, and which thoughts are merely personal thoughts and therefore should be ignored by the system?
More importantly, asks Dr. Mehrdad Fatourechi, can the system provide the users with the ability to control a computer whenever they want? These are the questions that Fatourechi and other “self-paced” brain computer interface (BCI) researchers are trying to answer.

So far, no self-paced BCI system has performed well enough to be suitable for practical applications. But Fatourechi, along with Professors Dr. Rabab K. Ward and Dr. Gary E. Birch from the University of British Columbia, Canada, have recently made a significant improvement with the development of a self-paced, fully automated brain-computer interface. The group’s results are published in a recent issue of the Journal of Neural Engineering. . . . [ read more Self Paced Brain Computer Interface Gets Closer to Reality ]

This could make a huge change in an aging population. No more will you have to be feeble and helpless in old age.

Capturing the spirit of that inspirational catch phrase, a university team lead by Professor Shigeki Toyama formally announced the development of a robot suit they hope will dramatically ease the burden on agricultural field workers as well as other manual labor intensive jobs.

Developed at the Tokyo University of Agriculture and Technology graduate school, with support from the Japan Ministry of Agriculture, Forestry, and Fisheries, the new robot suit attempts to address the challenges faced by Japan’s aging, and difficult to replace, rural farm population. . . [ read more New Robot Suit to Assist Japanese Farmers]