Future of Artificial Intelligence

The lack of a rigorous mathematical foundation for electronics impelled engineer Leon Chua to develop one, which led to the formulation of the memristor--a theoretical fourth basic circuit element in addition to the resistor, capacitor, and inductor where electric charge and magnetic flux come together. Since then the creation of memristors has been achieved, and their novel abilities might unlock key insights about the human brain that would be a tremendous step forward for the field of artificial intelligence. Advantages of memristors include rapid, nanosecond writing of data using a very small amount of energy, and retention of memristive memory even when the power is turned off. The most immediate potential application for memristors is as a flash memory replacement, while durability improvements should make memristors ideal for a superfast random access memory, says Hewlett-Packard (HP) Laboratories Fellow Stan Williams. The discovery that a slime mold was behaving in the manner of a memristive circuit in that it could memorize a pattern of events without the aid of a neuron inspired a physicist at the University of California, San Diego to construct a circuit capable of learning and predicting future signals. Much earlier, Chua had noticed a sharp similarity between synapse behavior and memristor response, leading to speculation that memristors might help engineer an electronic intelligence that can mimic the power of a brain. The U.S. Defense Advanced Research Projects Agency has embarked on a project to create "electronic neuromorphic machine technology that is scalable to biological levels." HP's Greg Snider has envisioned the field of cortical computing that focuses on the potential of memristors to imitate the interaction of the brain's neurons. He and Williams are working with Boston University scientists to devise hybrid transistor-memristor chips that aim to replicate some of the brain's thought processes.

Mind-Reading Computers

Washington University in St. Louis researchers have developed technology that gives computers the ability to understand speech imagined in the mind. "The idea is to basically connect people with devices and machines through their thoughts directly," says Washington University Medical School neurologist Eric Leuthardt. The research is based on brain-computer interface (BCI) technology, which monitors brainwaves and uses computers to decode those signals and transform them into action. So far, BCI research efforts have only been able to decode imagined actions. The ability to decode imagined speech will make mentally communicating with computers far easier. Leuthardt says the technology will better connect humans and machines, and will give the disabled unprecedented access to the world. Leuthardt and Washington University biological engineer Daniel Moran have developed video games that can be played with the mind. Players control the game by imagining an action. For example, imagining moving the left hand may mean moving left, while imagining moving the tongue may create upward movement. The system has only been tested on a few people because the sensors used require brain surgery. So far, children with epilepsy have been given the chance to participate because they already have similar equipment surgically implanted to locate electric signals in the brain.

Cars Driving Themselves

University of New Brunswick researchers led by professor Howard Li are developing cars capable of driving themselves. Li says one of the first steps is to take detailed pictures of sharp turns, deer, pedestrians, and other obstacles and program those images into a simulator so the system learns what objects to avoid. He says the biggest challenge is developing the right algorithms to allow thousands of smart cars to be compatible on the road and avoid collisions with each other. "We obviously can't use thousands of vehicles to test artificial intelligence compatibility," he says. "We'll use computer simulations to test it and write computer simulations of multiple cars working together." Li says the technology is likely decades away from being used in commercial cars. However, he says there will be breakthroughs made as part of the process that could lead to sensors that prevent accidents by warning drivers of potential dangers. The technology also could be used to save lives in Afghanistan by helping troops avoid hazards. Li says artificial intelligence technology could be used to automate vehicles for tasks such as snow removal, city transit, assembly lines, and farming. "The robotics market is growing 40 percent every year," he says. "This is a field that's going to keep growing and evolving, and one day it will be as common for every family as owning a PC is today."

Schools and Video Games

Teachers are increasingly incorporating video games, virtual reality, and simulations to improve education. Business and science classes are starting to use sophisticated software that allows students to test out potential careers, practice skills, or explore history through simulated adventures in national parks, ancient cities, or outer space. The military and medical schools, which use games and simulations to train new personnel, are helping to boost the use of video games in classrooms. Advocates argue that games can teach vital skills such as teamwork, decision-making, and digital literacy. Games also can challenge students just enough to keep them interested in reaching the next level. "There is a revolution in the understanding of the educational community that video games have a lot of what we need," says Jan Plass, co-director of the Games for Learning Institute at New York University. Game designers are replacing the violence in video games with equations and educational challenges. For example, Dimension M is a suite of math games that require players to learn about functions and solve equations to stop a biodigital virus from taking over the world. The Federation of American Scientists is promoting games as a way of inspiring new scientists, and has developed two games in which players fight bacterial invaders in a blood vessel. A recent revision to the Higher Education Act authorized the creation of a research center for assessing and developing educational technologies such as simulations and video games.

Educational Technology

University of Arizona (UA) scientists have received a $300,000 Defense Advanced Research Projects Agency grant to develop artificial intelligence and education technology that mirrors the consumer tracking algorithms used by sites such as Amazon and iTunes. The researchers, led by UA professor Paul R. Cohen, want to maximize a tutoring system model by using data on learners to improve the feedback provided by intelligent tutoring systems. "Teaching people means making a sequence of dependent decisions," Cohen says. "We're trying to optimize the value of each decision by reasoning algorithmically about how it sets up the student for future learning opportunities." The researchers are developing a program that would be capable of already knowing what a student knows and matching that knowledge with comparable students before suggesting specific texts, exams, videos, educational games, demonstrations, and other Web-based educational tools. The technology would be able to direct students to the best possible learning experience for each student, refining each student's curriculum as it learns more about learners in general. Preliminary findings from a pilot project suggest that students who used the model learned more quickly and were able to retain information better, and that the program improved as more students participated.

Mobile: Future of the Internet

Mobile phones will become the primary Internet device for most people by 2020, largely due to their increasing computing power, predicts the "Future of the Internet," a new Pew Internet & American Life Project report. "Telephony [will be] offered under a set of universal standards and protocols accepted by most operators internationally, making for reasonably effortless movement from one part of the world to another," the report says. The report, based on a survey of 578 Internet activists, builders, and commentators, also predicts that despite the widespread access to other cultures and viewpoints on the Internet, Internet use will not make people more socially tolerant. The report says that some survey respondents even suggested that the divide between the tolerant and the intolerant could widen due to Internet-based information-sharing tactics. The report also found that 55 percent of experts believe that by 2020 people will routinely interact in artificial spaces through virtual worlds and other types of augmented reality. Voice activation and touch interfaces will be common by 2020, predict nearly two-thirds of experts, and "air-typing" will become common because small handheld devices will display a full size virtual keyboard on any flat surface. A majority of experts, 78 percent, believe that the current Internet architecture will not be completely replaced by a new system in 2020, but search, security, and reliability will have been improved by next-generation research.

Microsoft Research Projects

Microsoft is supporting a host of projects in its research and development division that are focused on tackling real-world technical challenges but could have a potentially dramatic impact on computing. One such project is Eagle 1, a tool for disaster recovery teams that captures information from multiple databases and generates real-time interactive maps through the use of geospatial mapping technology. Microsoft Surface, which has reached the product phase, promotes social interaction through a multitouch table with a ruggedized acrylic interface. Its capabilities include object recognition, optical tagging, and interaction with physical objects. LucidTouch V2 technology is a credit card-sized display screen that the user can reach behind to control a mobile device. Visual Studio 2010 upgrades Microsoft's Visual Studio development platform with new features that include unified modeling language, a debugging tool that can pinpoint non-reproducible bugs by automatically generating data sets, and an application that lets development teams visualize a model of the existing development architecture and find any existing code assets that are not well categorized. The Touch Wall is a new hardware/software interface with multitouch control that can mix and match media on the same large-screen display to enhance collaboration. The OSLO project offers a framework that allows all members of a team to access data models in a repository across the entire software development life cycle. Microsoft also is working on a robotic receptionist to be installed at the company headquarters that will help visitors find transportation, using voice and facial recognition technology. Finally, BlueTrack is mouse technology that employs a wider and brighter laser beam than conventional optics so that it can operate on uneven surfaces.