New Cognitive Robotics Lab Tests Theories of Human Thought

"The real world has a lot of inconsistency that humans handle almost without noticing -- for example, we walk on uneven terrain, we see in shifting light," said Professor Vladislav Daniel Veksler, who is currently teaching Cognitive Robotics."With robots, we can see the problems humans face when navigating their environment."

Cognitive Robotics marries the study of cognitive science -- how the brain represents and transforms information -- with the challenges of a physical environment. Advances in cognitive robotics transfer to artificial intelligence, which seeks to develop more efficient computer systems patterned on the versatility of human thought.

Professor Bram Van Heuveln, who organized the lab, said cognitive scientists have developed a suite of elements -- perception/action, planning, reasoning, memory, decision-making -- that are believed to constitute human thought. When properly modeled and connected, those elements are capable of solving complex problems without the raw power required by precise mathematical computations.

"Suppose we wanted to build a robot to catch fly balls in an outfield. There are two approaches: one uses a lot of calculations -- Newton's law, mechanics, trigonometry, calculus -- to get the robot to be in the right spot at the right time," said Van Heuveln."But that's not the way humans do it. We just keep moving toward the ball. It's a very simple solution that doesn't involve a lot of computation but it gets the job done."

Robotics are an ideal testing ground for that principle because robots act in the real world, and a correct cognitive solution will withstand the unexpected variables presented by the real world.

"The physical world can help us to drive science because it's different from any simulated world we could come up with -- the camera shakes, the motors slip, there's friction, the light changes," Veksler said."This platform -- robotics -- allows us to see that you can't rely on calculations. You have to be adaptive."

The lab is open to all students at Rensselaer. In its first semester, the lab has largely attracted computer science and cognitive science students enrolled in a Cognitive Robotics course taught by Veksler, but Veksler and Van Heuveln hope it will attract more engineering and art students as word of the facility spreads.

"We want different students together in one space -- a place where we can bring the different disciplines and perspectives together," said Van Heuveln."I would like students to use this space for independent research: they come up with the research project, they say 'let's look at this.'"

The lab is equipped with five"Create" robots -- essentially a Roomba robotic vacuum cleaner paired with a laptop; three hand-eye systems; one Chiara (which looks like a large metal crab); and 10 LEGO robots paired with the Sony Handy Board robotic controller.

On a recent day, Jacqui Brunelli and Benno Lee were working on their robot"cat" and"mouse" pair, which try to chase and evade each other respectively; Shane Reilly was improving the computer"vision" of his robotic arm; and Ben Ball was programming his robot to maintain a fixed distance from a pink object waved in front of its"eye."

"The thing that I've learned is that the sensor data isn't exact -- what it 'sees' constantly changes by a few pixels -- and to try to go by that isn't going to work," said Ball, a junior and student of computer science and physics.

Ball said he is trying to pattern his robot on a more human approach.

"We don't just look at an object and walk toward it. We check our position, adjusting our course," Ball said."I need to devise an iterative approach where the robot looks at something, then moves, then looks again to check its results."

The work of the students, who program their robots with the Tekkotsu open-source software, could be applied in future projects, said Van Heuveln.

"As a cognitive scientist, I want this to be built on elements that are cognitively plausible and that are recyclable -- parts of cognition that I can apply to other solutions as well," said Van Heuveln."To me, that's a heck of a lot more interesting than the computational solution."

In a generic domain, their early investigations clearly show how a more cognitive approach employing limited resources can easily outpace more powerful computers using a brute force approach, said Veksler.

"We look to humans not just because we want to simulate what we do, which is an interesting problem in itself, but also because we're smart," said Veksler."Some of the things we have, like limited working memory -- which may seem like a bad thing -- are actually optimal for solving problems in our environment. If you remembered everything, how would you know what's important?"

Source

New Psychology Theory Enables Computers to Mimic Human Creativity

Solving this"insight problem" requires creativity, a skill at which humans excel (the coin is a fake --"B.C." and Arabic numerals did not exist at the time) and computers do not. Now, a new explanation of how humans solve problems creatively -- including the mathematical formulations for facilitating the incorporation of the theory in artificial intelligence programs -- provides a roadmap to building systems that perform like humans at the task.

Ron Sun, Rensselaer Polytechnic Institute professor of cognitive science, said the new"Explicit-Implicit Interaction Theory," recently introduced in an article inPsychological Review, could be used for future artificial intelligence.

"As a psychological theory, this theory pushes forward the field of research on creative problem solving and offers an explanation of the human mind and how we solve problems creatively," Sun said."But this model can also be used as the basis for creating future artificial intelligence programs that are good at solving problems creatively."

The paper, titled"Incubation, Insight, and Creative Problem Solving: A Unified Theory and a Connectionist Model," by Sun and Sèbastien Hèlie of University of California, Santa Barbara, appeared in the July edition ofPsychological Review. Discussion of the theory is accompanied by mathematical specifications for the"CLARION" cognitive architecture -- a computer program developed by Sun's research group to act like a cognitive system -- as well as successful computer simulations of the theory.

In the paper, Sun and Hèlie compare the performance of the CLARION model using"Explicit-Implicit Interaction" theory with results from previous human trials -- including tests involving the coin question -- and found results to be nearly identical in several aspects of problem solving.

In the tests involving the coin question, human subjects were given a chance to respond after being interrupted either to discuss their thought process or to work on an unrelated task. In that experiment, 35.6 percent of participants answered correctly after discussing their thinking, while 45.8 percent of participants answered correctly after working on another task.

In 5,000 runs of the CLARION program set for similar interruptions, CLARION answered correctly 35.3 percent of the time in the first instance, and 45.3 percent of the time in the second instance.

"The simulation data matches the human data very well," said Sun.

Explicit-Implicit Interaction theory is the most recent advance on a well-regarded outline of creative problem solving known as"Stage Decomposition," developed by Graham Wallas in his seminal 1926 book"The Art of Thought." According to stage decomposition, humans go through four stages -- preparation, incubation, insight (illumination), and verification -- in solving problems creatively.

Building on Wallas' work, several disparate theories have since been advanced to explain the specific processes used by the human mind during the stages of incubation and insight. Competing theories propose that incubation -- a period away from deliberative work -- is a time of recovery from fatigue of deliberative work, an opportunity for the mind to work unconsciously on the problem, a time during which the mind discards false assumptions, or a time in which solutions to similar problems are retrieved from memory, among other ideas.

Each theory can be represented mathematically in artificial intelligence models. However, most models choose between theories rather than seeking to incorporate multiple theories and therefore they are fragmentary at best.

Sun and Hèlie's Explicit-Implicit Interaction (EII) theory integrates several of the competing theories into a larger equation.

"EII unifies a lot of fragmentary pre-existing theories," Sun said."These pre-existing theories only account for some aspects of creative problem solving, but not in a unified way. EII unifies those fragments and provides a more coherent, more complete theory."

The basic principles of EII propose the coexistence of two different types of knowledge and processing: explicit and implicit. Explicit knowledge is easier to access and verbalize, can be rendered symbolically, and requires more attention to process. Implicit knowledge is relatively inaccessible, harder to verbalize, and is more vague and requires less attention to process.

In solving a problem, explicit knowledge could be the knowledge used in reasoning, deliberately thinking through different options, while implicit knowledge is the intuition that gives rise to a solution suddenly. Both types of knowledge are involved simultaneously to solve a problem and reinforce each other in the process. By including this principle in each step, Sun was able to achieve a successful system.

"This tells us how creative problem solving may emerge from the interaction of explicit and implicit cognitive processes; why both types of processes are necessary for creative problem solving, as well as in many other psychological domains and functionalities," said Sun.

Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of ScienceDaily or its staff.

Source

Breakthrough Chip Technology Lights Path to Exascale Computing: Optical Signals Connect Chips Together Faster and With Lower Power

The new technology, called CMOS Integrated Silicon Nanophotonics, is the result of a decade of development at IBM's global research laboratories. The patented technology will change and improve the way computer chips communicate -- by integrating optical devices and functions directly onto a silicon chip, enabling over 10X improvement in integration density than is feasible with current manufacturing techniques.

IBM anticipates that Silicon Nanophotonics will dramatically increase the speed and performance between chips, and further the company's ambitious exascale computing program, which is aimed at developing a supercomputer that can perform one million trillion calculations -- or an exaflop -- in a single second. An exascale supercomputer will be approximately one thousand times faster than the fastest machine today.

"The development of the Silicon Nanophotonics technology brings the vision of on-chip optical interconnections much closer to reality," said Dr. T.C. Chen, vice president, Science and Technology, IBM Research."With optical communications embedded into the processor chips, the prospect of building power-efficient computer systems with performance at the exaflop level is one step closer to reality."

In addition to combining electrical and optical devices on a single chip, the new IBM technology can be produced on the front-end of a standard CMOS manufacturing line and requires no new or special tooling. With this approach, silicon transistors can share the same silicon layer with silicon nanophotonics devices. To make this approach possible, IBM researchers have developed a suite of integrated ultra-compact active and passive silicon nanophotonics devices that are all scaled down to the diffraction limit -- the smallest size that dielectric optics can afford.

"Our CMOS Integrated Nanophotonics breakthrough promises unprecedented increases in silicon chip function and performance via ubiquitous low-power optical communications between racks, modules, chips or even within a single chip itself," said Dr. Yurii A. Vlasov, Manager of the Silicon Nanophotonics Department at IBM Research."The next step in this advancement is to establishing manufacturability of this process in a commercial foundry using IBM deeply scaled CMOS processes."

By adding just a few more processing modules to a standard CMOS fabrication flow, the technology enables a variety of silicon nanophotonics components, such as: modulators, germanium photodetectors and ultra-compact wavelength-division multiplexers to be integrated with high-performance analog and digital CMOS circuitry. As a result, single-chip optical communications transceivers can now be manufactured in a standard CMOS foundry, rather than assembled from multiple parts made with expensive compound semiconductor technology.

The density of optical and electrical integration demonstrated by IBM's new technology is unprecedented -- a single transceiver channel with all accompanying optical and electrical circuitry occupies only 0.5mm²-- 10 times smaller than previously announced by others. The technology is amenable for building single-chip transceivers with area as small as 4x4mm²that can receive and transmit over Terabits per second that is over a trillion bits per second.

The development of CMOS Integrated Silicon Nanophotonics is the culmination of a series of related advancements by IBM Research that resulted in the development of deeply scaled front-end integrated Nanophotonics components for optical communications. These milestones include:

• March 2010, IBM announced a Germanium Avalanche Photodetector working at unprecedented 40Gb/s with CMOS compatible voltages as low as 1.5V. This was the last piece of the puzzle that completes the prior development of the"nanophotonics toolbox" of devices necessary to build the on-chip interconnects.
• March 2008, IBM scientists announced the world's tiniest nanophotonic switch for"directing traffic" in on-chip optical communications, ensuring that optical messages can be efficiently routed.
• December 2007, IBM scientists announced the development of an ultra-compact silicon electro-optic modulator, which converts electrical signals into the light pulses, a prerequisite for enabling on-chip optical communications.
• December 2006, IBM scientists demonstrated silicon nanophotonic delay line that was used to buffer over a byte of information encoded in optical pulses -- a requirement for building optical buffers for on-chip optical communications.

The details and results of this research effort was reported in a presentation delivered by Dr. Yurii Vlasov at the major international semiconductor industry conference SEMICON held in Tokyo on Dec. 1, 2010. The talk is entitled"CMOS Integrated Silicon Nanophotonics: Enabling Technology for Exascale Computational Systems," co-authored by William Green, Solomon Assefa, Alexander Rylyakov, Clint Schow, Folkert Horst, and Yurii Vlasov of IBM's T.J. Watson Research Center in Yorktown Heights, N.Y. and IBM Zurich Research Lab in Rueschlikon, Switzerland.

Additional information on the project can be found athttp://www.research.ibm.com/photonics.

Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.

Source