• Also see: Lightwave secures funding. Ireland On-Line (August 17, 2004). "Lightwave Technologies, an innovative environmental technology company located at NovaUCD, has secured seed capital funding of ¤300,000 from a group of private investors. Lightwave Technologies uses artificial intelligence techniques to make efficient decisions for controlling energy usage in commercial buildings with the objective of saving up to 30% of energy costs for clients. ... This system is using new advances in computer science in the area of neural networks."

August 12, 2004: When machines breed - Evolvable hardware -- gadgets that design themselves -- can get the job done, even if humans have no idea how they do it. By Sam Williams. Salon.com (subscribe or watch a brief ad to get a free day pass). "Paul Layzell is a specialist in the budding field of evolvable hardware. Simply put, he helps machines design themselves, using principles borrowed directly from biological evolution. ... Using evolutionary processes to optimize machine performance is nothing new. Since the 1960s, artificial intelligence researchers have exploited the dynamics of Darwinian evolution to solve software problems in fields as diverse as financial investment, manufacturing and biochemistry. What is new, however, is the application of evolutionary processes in the hardware realm."
>>> Genetic Algorithms, Engineering, Machine Learning, Finance & Investing, Manufacturing, Bioinformatics, Applications

August 3, 2004: Neural network mimics human thought process. HP interested in technology invented at U of I. By Julie Howard. The Idaho Statesman. "Creation of a machine that thinks like a person is a step closer to reality with a discovery from a team of researchers at the University of Idaho in Moscow. The new technology, with a patent pending, 'opens the door' for computers or robots to do rapid computations in a much more complex way than scientists had thought possible, said inventor Richard Wells. ... The technology is called a neural network, and it operates much as the human brain does, Wells said. Instead of a microprocessor, which performs computation after computation, a neural network can do several computations simultaneously. In addition, a 'neurocomputer' uses what's called 'neuro-fuzzy logic,' meaning it deals with uncertainty, the missing function of traditional programmed integrated circuitry. ... Having a neural network formed on a computer chip could revolutionize the way computers function, said Gene Merrell, acting director of the Idaho Research Foundation Inc., based at the university."
>>> Neural Networks, Fuzzy Logic, Machine Learning, Systems

August 2004: A Machine With a Mind of Its Own - Ross King wanted a research assistant who would work 24/7 without sleep or food. So he built one. By Oliver Morton. Wired Magazine (Issue 12.08). "For a machine that's changing the world, the device on the lab bench in front of me doesn't look very impressive - it just goes back and forth, back and forth, back and forth. ... [Ross] King's humble robot is based on a Biomek 2000, a low-rent fluid-handling device that goes for only $37,900. But it can do something its more nimble cousins can't. Its components - the tireless robot arm, an incubator in which cells cultured on the platter either wither or thrive, and a plate reader that examines the little depressions to see whether anything is growing there - are linked up to a much more exceptional brain. The artificial intelligence routines in that brain can look at the results of an experiment, draw a conclusion about what the results might mean, and then set off to test that conclusion. The 'robot scientist' (King has resisted the temptation of a jazzy acronym) may look like a mere labor-saving gizmo, shuttling back and forth ad nauseam, but it's much more than that. Biology is full of tools with which to make discoveries. Here's a tool that can make discoveries on its own. ... Studying AI at the Turing Institute in Glasgow, [King] set about using machine-learning techniques to predict the shapes of proteins, one of the fundamental challenges of bioinformatics. King, though, found a twist. With his friend Colin Angus, whom he'd met at Aberdeen, he developed software that translated protein structures into musical chord sequences.... Stephen Muggleton argues that the life sciences are peculiarly well suited to machine learning. 'There's an inherent structure in biological problems that lends itself to computational approaches,' he says. In other words, biology reveals the machinelike substructure of the living world; it's not surprising that machines are showing an aptitude for it."
>>> Bioinformatics, Machine Learning, Robots, Scientific Discovery, Applications, Music

July / August 2004: Spotting Cancer Sooner - Blood tests that detect cancer in its early stages would save countless lives. The first could arrive within a year. By Ken Garber. Technology Review. "The individual fates of the 1.3 million Americans diagnosed with cancer this year will be largely decided by one simple factor: at what stage was the disease spotted? ... The problem, of course, is that cancers, which begin with just a few deviant cells, are by their very nature hard to diagnose early. In the last few years, though, a new method has emerged that promises to deliver simple blood tests that identify the telltale molecular profiles of various cancers easily and accurately. ... Like [George] Wright, [Emanuel] Petricoin and [Lance] Liotta used a Ciphergen system to generate protein profiles from blood samples. Their early attempts to find cancer patterns failed, though, because they were simply trying to juggle too much information. Then, in June 1999, a solution appeared. Petricoin and his friend Peter Levine, a Maryland lawyer with a background in data analysis, were chatting about the problem over brunch; Levine suggested using pattern recognition algorithms to make sense of the massive amount of data. Levine, who had considered using such algorithms to analyze stock market trends and commodities trading, sketched out the cancer idea on a napkin. 'In about five minutes, we both realized this would be a really fascinating approach,' Petricoin recalls. So they tested it, together with Ben Hitt, a software engineer who borrowed the necessary algorithms from artificial-intelligence theory. In fact, cancer patterns did emerge, and in 2000 Levine and Hitt founded Correlogic Systems to develop blood tests for cancers. In early 2002, the researchers published results in the British medical journal Lancet , showing they could use a specific protein pattern to spot ovarian cancer."
>>> Medicine, Pattern Recognition, Finance & Investing, Machine Learning, Applications

July 29, 2004: Organic PC goal of UK project. By Harry Yeates. Electronics Weekly. "In the future, alongside the box of lifeless silicon you call your PC, you might find a little tub of living tissue. For particular specialist tasks involving complex, non-linear problems your inorganic circuits would find daunting, you would turn to the box of organics. That's the ultimate aim of a new £1.2m, four year research project involving the universities of the West of England (UWE), Leeds and Sussex. 'For fifty years AI has been trying to build systems that have got complicated behaviour, with some success,' said Dr Larry Bull from UWE, who will lead the project. 'But given this complex behaviour seems to be easy in the natural world, networks of neurons and chemical systems, why don't we try to build AI systems out of that stuff, rather than try to write clever programmes?'"
>>> Systems, Neural Networks, Machine Learning

July 28, 2004: Amplified Intelligence - The AI Problem. Interview with Ken Ford. Astrobiology Magazine. "Astrobiology Magazine (AM): The IMHC [Interdisciplinary Study of Human & Machine Cognition] research agenda broadly seems to cover robotics, cognition and simulations. Are there parts of machine intelligence that your research institute doesn't cover today, but that you see as growth areas? Ken Ford (KF): Don't forget that second letter is 'H'. Although a lot of our research could be categorized as AI, and five of our researchers are AAAI (American Association for Artificial Intelligence) Fellows, IHMC is not a traditional machine intelligence laboratory. The focus and theme of our research is what has become known as human-centered computing which, in a nutshell, is about fitting technology to people instead of fitting people to technology. The human is part of the system, and it is the performance of the whole system, including the human, that we are interested in. This requires that machines should be designed to fit us physically, cognitively, and perhaps even socially. We think of AI as meaning 'Amplified Intelligence.' The interesting thing is that many traditional AI technologies in fact are being used in just this way. We like to refer to it as building cognitive prostheses, computational systems that leverage and extend human intellectual capacities, just as eyeglasses are a kind of ocular prosthesis. Building cognitive prostheses is fundamentally different from AI's traditional Turing Test ambitions -- it doesn't set out to imitate human abilities, but to extend them. ... AM: In your opinion, how well do the machine intelligence problems (like navigation, data-mining, or simulations with agents) map to the basic computer science [CS] problem of efficient 'search'? KF: Wow, efficient search is a 'basic computer science problem'? Not long ago, search was being suggested as a defining characteristic of AI to distinguish it from 'mainstream' CS. But to return to the question: search is certainly a central technique in AI, but the search spaces arising in AI are often impossibly huge, and a more interesting aspect is not so much how to search them efficiently as how to re-cast problems so that the search space itself is reduced in size. Searching is what you do when you can't think of anything smarter."
>>> Interfaces, Search, Robots, Space Exploration, Data Mining, Household Appliances, Interviews, AI Overview, Applications, Reasoning, Machine Learning, Turing Test

July 16, 2004: I.T. May Help Clean a Polluted Sea, Say Researchers. By Mike Martin. NewsFactor Network. "If an article in this week's journal Science is on target, air pollution fouls not only our skies but our oceans as well. ... But software and information technology may play an equally important role, claim the authors of a study published in a recent special issue of the journal Management of Environmental Quality, which is devoted to 'information technologies in environmental engineering.' 'Rapid environmental changes call for continuous surveillance and online decision-making -- two areas where I.T. can be valuable,' say study authors Ioannis Athanasiadis and Pericles Mitkas. Both are computer science researchers at the Informatics and Telematics Institute Center for Research and Technology in Thessaloniki, Greece. In their study, entitled 'An Agent-Based Intelligent Environmental Monitoring System,' the researchers 'present a multi-agent system for monitoring and assessing air-quality attributes, which uses data coming from a meteorological station.' Their system, the study explains, uses a 'community of software agents to monitor and validate measurements coming from several sensors to assess air-quality.' Software agents are computer systems to which an operator can delegate tasks. Like the robots in the new movie 'I, Robot,' software agents are more autonomous, proactive and adaptive than the everyday software we normally use. ... Using agents to monitor the environment is a branch of 'enviromatics -- the research initiative examining the application of information technology in environmental research, monitoring, assessment, management and policy,' Athanasiadis explains. ... 'In O3RTAA, several software agents operate in a distributed-agent society in order to monitor both meteorological and air pollutants, to evaluate air quality and, ultimately, to trigger alarms' about environmental damage, Mitkas explains, adding that the system uses machine-learning algorithms and data-mining methodologies for 'extracting knowledge.'"
>>> Multi-Agent Systems, Data Mining, Natural Resource Management & The Environment, Agents, Machine Learning, Applications

July 16, 2004: Movie tests Asimov's moral code for robots. By Will Knight. New Scientist News. "The possibility of developing truly intelligent machines, and their potential to be friend or foe to humanity, gets the Hollywood treatment in a new blockbuster film I, Robot, which opens in the US on Friday. At the heart of the movie are Isaac Asimov's 'Three Laws of Robotics', invented as a simple, but immutable moral code for robots. ... [R]obotics and artificial intelligence experts admit they are a long way from having to worry about such rules yet. 'The difficulty is building something that would understand them,' says Alan Bundy, at Edinburgh University's Artificial Intelligence Institute in the UK. 'That is well beyond the state of the art at the moment.' Bundy notes that simple safety measures are already a crucial part of the design of industrial robots, which have in rare cases caused the death of people. ... 'Asimov's laws are about as relevant to robotics as leeches are to modern medicine,' says Steve Grand, who founded the UK company Cyberlife Research and is working on developing artificial intelligence through learning. 'They stem from an innocent bygone age, when people seriously thought that intelligence was something that could be 'programmed in' as a series of logical propositions.'"
>>> Robots, Science Fiction, Ethical & Social Implications, Machine Learning, Manufacturing

July 14, 2004: Computer brains. e4engineering.com. "A team of computer scientists and mathematicians at Palo Alto, CA-based Artificial Development are developing software to simulate the human brain's cortex and peripheral systems. As a first step along the way, the company recently disclosed that it has completed the development a realistic representation of the workflow of a functioning human cortex. Dubbed the CCortex-based Autonomous Cognitive Model ('ACM'), the software may have immediate applications for data mining, network security, search engine technologies and natural language processing."
>>> Neural Networks & Connectionist Systems, Machine Learning, Natural Language Processing, Cognitive Science, Data Mining, Information Retrieval, Networks, Applications

July 14, 2004: Attack of the killer vacuum cleaners. By Charles Arthur. The Belfast Telegraph Digital. "Things are about to happen with robots, because the element they need to make them truly useful - the software, which needs to be able to adapt to a wide range of situations - is getting cheaper all the time. Future Horizons, a semiconductor analyst based in Kent, forecasts that by 2010 there will be 55.5 million robots, in a world market worth £30bn - up from £2.4bn last year. 'The electronics industry is on the cusp of a robotics wave, a period in which applications are aimed at labour-saving and extending human skills,' it reports. Of those, it says that 39 million will be domestic robots, and 10.5 million 'domestic intelligent service' robots. That is because there's a growing need for robots to help the elderly and handicapped. ... But the real explosion in robotics is coming among the 'immobots' - or, more simply, just 'bots'. These are bits of software that are incorporated into larger objects, and that remove a lot of the strain of having to decide what to do next. We're getting glimpses of how good these could be at present: the tiny number of Britons with a TiVo personal video recorder have something that decides, based on the programmes they choose to record, what other programmes they might like to see, and records those, too. ... The reason why we can't yet declare 'The Year of the Robot', however, is that researchers are still fundamentally split about how robots should behave and learn. One group favours the 'top-down' approach, in which all the behaviour of the robot is mapped out, and its software is written to fill out that behaviour. The Roomba vacuum cleaner is a classic example.... The alternative is something assembled from smaller, self-contained units, which creates a gestalt of behaviour based on that. Thus the system that controls the legs learns to 'walk' independently.... Sony's Aibo draws on a form of this....
>>> Robots, Science Fiction, Agents, Systems, Assisitive Technologies, Household Appliances, Industry Statistics, Applications, Reasoning, Machine Learning

July 7, 2004: Software aids future tennis stars. BBC News. " As Britons bemoan another year without a Wimbledon hero, there could be some hope in a computer model being worked on at Kingston University in London. ... It will create a computer-generated competitor which rival players can pit themselves against. The system will analyse video footage of champions and allow other players to explore tactics to beat them. ... The research will focus initially on tennis but will move on to look at more complex sports such as football and basketball. 'As well as helping specialised sports training, the technology we are developing could have benefits in fields such as realistic computer gaming, virtual reality and surveillance,' said Dr Ahmed Shihab of the School of Computing and Information Systems at Kingston University."
>>> Sports, Pattern Recognition, Vision, Machine Learning, Video Games, Law Enforcement, Information Retrieval, Applications

July 6, 2004: Programmer seems to have a technology that does everything. By Rachel Melcer. St. Louis Post-Dispatch / STLtoday.com. "Steven Thaler, founder of Imagination Engines Inc. in Maryland Heights, says he has a unique challenge: figuring out what to do with a technology that does everything. He and his supporters say his creation, a computer program called the Creativity Machine, has huge economic potential. It could be the first successful form of artificial intelligence, a machine that learns and thinks by simulating the human brain's activity. ... Imagination Engines also is experimenting with spinoff companies that license the core technology and adapt it for specific uses. ... The first spinoff, Synaptrix Financial Prediction LLC, was created last year as a partner for Stann Financial. It aims to analyze a real-time flow of information on trades in the financial markets to predict the best time to buy or sell a particular stock. The project showed early promise, reaching a 60 percent to 65 percent accuracy rate, but it stalled over problems with the information feed and the need to refine its programming, [John] Stann said. ... Synaptrix Parts Inspection LLC, another of his spinoffs, combines an ordinary video camera with the Creativity Machine's neural network and custom software to perform quality-control checks in manufacturing. The system is 'shown' a variety of objects that it can learn to instantly identify for sorting or to use as an ideal to spot defects and variations. ... On the government side, Imagination Engines is part of a consortium developing an airport-security system for the Department of Homeland Security. The group recently got an 18-month, $800,000 grant to design and test a series of smart sensors at an airport in Butte, Mont. The system would be able to identify vehicles on airport property, monitor them, spot and warn of suspicious activity, Thaler said."
>>> Neural Networks, Image Understanding, Finance & Investing, Manufacturing, Law Enforcement, Machine Learning, Vision, Applications

July 6, 2004: Evolution could speed net downloads. By Will Knight. New Scientist News. "Transferring popular data across the internet repeatedly can be inefficient and costly, so networking companies have developed ways of temporarily storing, or 'caching', data at different locations to reduce costs and increase download speeds. But figuring out where to store data and for how long is a complex problem. One solution might be to have caches 'talk' to each other repeatedly, but this is inefficient as it takes up a lot of bandwidth. To tackle the challenge, Pablo Funes of US company Icosystem and Jürgen Branke and Frederik Theil of the University of Karlsruhe in Germany used 'genetic algorithms', which mimic Darwinian evolution, to develop strategies for internet servers to use when caching data. Using a simulation they were able to improve download speeds over existing caching schemes. ... Funes told New Scientist the scheme could eventually be used to allow caches to automatically 'evolve' their configuration."
>>> Genetic Algorithms, Telecommunications, Networks, Machine Learning, Applications

July 4, 2004: His quest - Do Disney in a day. By Larry Bleiberg. The Dallas Morning News / available from Mickey News. "Rich Vosburgh worked out hard, spending four months with a personal trainer. He scrutinized maps and a detailed timetable. He even deployed a secret weapon: artificial-intelligence research to chart a course through death-defying drops, torrents of water and fiery heat.And when this Texas adventurer clambered out of a floating log a year ago, he had reached his holy grail: visiting - in a single day - each of the 41 operating rides, attractions and shows at the Everest of theme parks, Walt Disney World's Magic Kingdom. His time: a record 10 hours, 40 minutes. ... At heart, the challenge is an enduring and perplexing quandary: What's the most efficient way to route someone to multiple places, taking into account constantly changing conditions? Logistics and timing Mathematicians call it the Time Dependent Traveling Salesman Problem. The answer could help fighter-jet pilots chart bombing targets or freight companies schedule package deliveries."
>>> Traveling Salesperson Problem, Planning and Scheduling, Search, Genetic Algorithms, Machine Learning, Reasoning, Transportation, Applications, Games & Puzzles

July 2004 [issue date]: Homeland Security as Catalyst - Innovative software firms are answering the call from U.S. government agencies for advanced analytics to help combat terrorism and criminal activity. What's the potential of this software for strategic business applications? By Jesus Mena. Intelligent Enterprise Magazine. "Ever heard of NORA? Or how about these guys: InferAgent, CopLink, NameHunter, Bladeworks, and Sentinel? These ominous-sounding fellows are products from tiny software firms that are developing some of the most advanced analytic technologies today for homeland security. Some provide solutions for the conversion of garbled text into knowledge discovery. Others tend to the unearthing of associations of individuals to actions, locations, and events from hundreds of thousands of internal and external records. Still others offer innovative methods for detecting fraud, categorizing foreign names, and virtual, remote analysis of data or text from any database in the world for agencies such as the U.S. Department of Homeland Security's Terrorist Threat Integration Center (TTIC). Given the growing diversity and globalization of business enterprises, is it possible that these innovative technologies, finding clear purpose for homeland security, could also be of interest to private business enterprises? In this article, I will describe some of these new technologies and how they may be applied to your company today and tomorrow. Who Are These Guys? Innovative products I mentioned at the beginning are commercial off-the-shelf (COTS) software -- a term favored by military and government agencies -- originating from such companies as Attensity, InferX, Infoglide, Knowledge Computing Corp. (KCC), Language Analysis Systems (LAS), Searchspace, System Research & Development (SRD), and others. Almost all have developed applications based on artificial intelligence technologies to meet demand from first military and intelligence communities, and now from the emerging homeland security market."
>>> Law Enforcement, Military, Information Extraction, Business, Knowledge Management, Applications, Machine Learning, Natural Language Processing, Agents

Spring 2004: What We Don't Know Can Hurt Us. By Heather Mac Donald. City Journal (Vol. 14, No. 2). "Immediately after 9/11, politicians and pundits slammed the Bush administration for failing to 'connect the dots' foreshadowing the attack. What a difference a little amnesia makes. For two years now, left- and right-wing advocates have shot down nearly every proposal to use intelligence more effectively -- to connect the dots -- as an assault on 'privacy.' Though their facts are often wrong and their arguments specious, they have come to dominate the national security debate virtually without challenge. The consequence has been devastating: just when the country should be unleashing its technological ingenuity to defend against future attacks, scientists stand irresolute, cowed into inaction. 'No one in the research and development community is putting together tools to make us safer,' says Lee Zeichner of Zeichner Risk Analytics, a risk consultancy firm, 'because they're afraid' of getting caught up in a privacy scandal. The chilling effect has been even stronger in government. 'Many perfectly legal things that could be done with data aren't being done, because people don't want to lose their jobs,' says a computer security entrepreneur who, like many interviewed for this article, was too fearful of the advocates to let his name appear. ... The goal of TIA [the Total Information Awareness project] was this: to prevent another attack on American soil by uncovering the electronic footprints terrorists leave as they plan and rehearse their assaults. ... TIA would have been the most advanced application yet of a young technology called 'data mining,' which attempts to make sense of the explosion of data in government, scientific, and commercial databases. Through complex algorithms, the technique can extract patterns or anomalies in data collections that a human analyst could not possibly discern. ... Without question, TIA represented a radical leap ahead in both data-mining technology and intelligence analysis, not surprising for a visionary group like DARPA, which created the Internet. ... As with any public or private power, TIA's capabilities could have been abused -- which is why DARPA planned to build safeguards throughout the system. But it differed from existing law enforcement and intelligence techniques only in degree, not kind. Though the scale of data it would have made immediately available to government was unprecedented, the type of evidence was identical to what government had had legal access to for decades. ... Information technology can help government in its constitutional responsibilities to protect the nation; indeed the congressional jo int inquiry into September 11 found that 'a reluctance to develop and implement new technical capabilities aggressively' was a cause of the pre-9/11 intelligence failures. The report added: 'While technology remains one of this nation's greatest advantages, it has not been fully and most effectively applied in support of U.S. counterterrorism efforts.' The privocrats will rightly tell you that eternal vigilance is the price of liberty; trouble is, they are aiming their vigilance at the wrong target." [Other projects discussed in this article: Human Identity at a Distance ; LifeLog; CAPPS II, Computer Assisted Passenger Prescreening System; MATRIX, Multistate Anti-Terrorism Information Exchange; and FIDNet.]
>>> Ethical & Social Implications, Data Mining, Law Enforcement, Military, Applications, Agents, Vision, Machine Learning

June 29, 2004: Panel Seeks Protections From Data Mining. By Brian Bergstein. Associated Press / available from The Herald News. "Even as the government increasingly relies on of data mining - scouring databases in search of clues about terrorism and everyday waste and fraud - there aren't clear rules about the practice. Privacy activists say it's like the wild West, dangerously unregulated. ... The data mining frontier could finally be seeing some civilizing influences take shape, particularly in the recommendations of a panel headed by former Federal Communications Commission chief Newton Minow that are getting particular praise. The panel's report, released in early June, acknowledged the importance of data mining in fighting terrorism. But it also said broad searches through reams of records and commercial files, on citizens who have done nothing to warrant individual suspicion, threaten fundamental protections in the Bill of Rights. To strike a balance, the group, known as the Technology and Privacy Advisory Committee (TAPAC), called for technological changes that would 'anonymize' data so investigators could hunt for suspicious activities and associations without immediately knowing whom they were probing."
>>> Ethical & Social Implications, Data Mining, Law Enforcement, Machine Learning, Applications

June 24, 2004: 2020 Vision has CCTV intelligent cameras deal in focus. The Journal / available from ic Newcastle. "Security specialist 2020 Vision Systems has secured an exclusive deal to provide artificial intelligence systems for CCTV cameras. The technology developed by Australian company, iOmniscient, allows security cameras to 'learn' to recognise anomalies in an area while ignoring routine movements. ... Using the technology, a camera can be 'taught' to recognise when valuable objects - such as paintings in a gallery - are moved, while ignoring people walking."
>>> Law Enforcement, Machine Learning, Applications, Vision

June 24, 2004: SkyNet Autonomy - Smart Satellite to Monitor Flood Gates. By Ed Stiles. University of Arizona report / available from Astrobiology Magazine. "There's nothing worse than a satellite that can't make decisions. Rather than organizing data, it simply spews out everything it collects, swamping scientists with huge amounts of information. It's like getting a newspaper with no headlines or section pages in which all the stories are strung together end-to-end. Researchers at the University of Arizona (UA), Arizona State University (ASU) and the Jet Propulsion Laboratory (JPL) are working to solve this problem by developing machine-learning and pattern-recognition software. This smart software can be used on all kinds of spacecraft, including orbiters, landers and rovers. Scientists currently are developing this kind of software for NASA's EO-1 satellite. The smart software allows the satellite to organize data so it sends back the most timely news first, while holding back less-timely data for later transmission. Although the project, called the Autonomous Sciencecraft Experiment (ASE), is still in the test and development stage, software created by UA hydrologists has already detected flooding on Australia's Diamantina River. ... The flood-detection software compares images from the satellite's cameras with images stored in its computer memory. If the rivers are not flooding and images come close to matching, the satellite remains silent. But if the satellite's computer finds significant differences, it takes more photos and notifies scientists."
>>> Space Exploration, Earth Science, Natural Resource Management, Pattern Recognition, Machine Learning, Applications

June 24, 2004: Informed decisions - CHEO team tests artificial intelligence in neo-natal unit. By Andrew Mayeda. Ottawa Citizen (subscription required). "When a baby is born prematurely, parents must often make a heartbreaking decision of whether to continue care or to simply let go. While that decision will never be easy, a pair of Ottawa researchers have developed artificial-intelligent tools that could at least make it more informed. The result is a software system [Parents Assisting Decision Support] that lets parents know their child's chances of survival, and allows them to weigh the pros and cons of treatment options while consulting their doctor or nurse. ... PADS, as it is called for short, is the brainchild of Dr. Robin Walker and Monique Frize, who have worked together for more than a decade."
>>> Case-Based Reasoning, Neural Networks, Medicine, Applications, Reasoning, Machine Learning

June 23, 2004: The Futurist - The Intelligent Internet. The Promise of Smart Computers and E-Commerce. By William E. Halal. Government Computer News Daily News (GCN). "Information and communication technologies are rapidly converging to create machines that understand us, do what we tell them to, and even anticipate our needs. We tend to think of intelligent systems as a distant possibility, but two relentless supertrends are moving this scenario toward near-term reality. Scientific advances are making it possible for people to talk to smart computers, while more enterprises are exploiting the commercial potential of the Internet. ... [F]orecasts conducted under the TechCast Project at George Washington University indicate that 20 commercial aspects of Internet use should reach 30% 'take-off' adoption levels during the second half of this decade to rejuvenate the economy. Meanwhile, the project's technology scanning finds that advances in speech recognition, artificial intelligence, powerful computers, virtual environments, and flat wall monitors are producing a 'conversational' human-machine interface. These powerful trends will drive the next generation of information technology into the mainstream by about 2010. ... The following are a few of the advances in speech recognition, artificial intelligence, powerful chips, virtual environments, and flat-screen wall monitors that are likely to produce this intelligent interface. ... IBM has a Super Human Speech Recognition Program to greatly improve accuracy, and in the next decade Microsoft's program is expected to reduce the error rate of speech recognition, matching human capabilities. ... MIT is planning to demonstrate their Project Oxygen, which features a voice-machine interface. ... Amtrak, Wells Fargo, Land's End, and many other organizations are replacing keypad-menu call centers with speech-recognition systems because they improve customer service and recover investment in a year or two. ... General Motors OnStar driver assistance system relies primarily on voice commands, with live staff for backup; the number of subscribers has grown from 200,000 to 2 million and is expected to increase by 1 million per year. The Lexus DVD Navigation System responds to over 100 commands and guides the driver with voice and visual directions. ... BCC Corporation estimates total AI sales to grow from $12 billion in 2002 to $21 billion in 2007. ... This scenario is not without uncertainties. Cynicism persists over unrealized promises of AI, and the Intelligent Internet will present its own problems. If you think today's dumb computers are frustrating, wait until you find yourself shouting at a virtual robot that repeatedly fails to grasp what you badly want it to do. ... The main obstacle is a lack of vision among industry leaders, customers, and the public as scars of the dot-com bust block creative thought."
>>> AI Overview, Applications, Natural Language Processing, Speech, Interfaces, Systems, Machine Learning, Customer Relations & E-Commerce, Information Retrieval, Networks, Industry Statistics

June 21, 2004: The Future of Business Intelligence & Predictions For BI's Future By Mitch Betts. Computerworld. [These articles are part of their special Business Intell igence report.] "We asked some industry leaders for their boldest predictions about the future of business intelligence tools, and here's our collection of the most interesting ideas. ... BI meets AI. In the near future, business leaders will manage by exception, and automated systems will handle significant loads of routine tasks. Today, automated systems in banking match incoming customer requests and inquiries with basic cross-sell and upsell oriented advertising. Over the next five years, these systems will become increasingly complex by considering customer financial status and wealth, transactional history, and even family and business relationships, to produce complex man/machine interactions that resemble artificial intelligence. The viability of artificial intelligence to solve real-world problems is being made possible by the convergence of hardware capabilities (faster processors, memory expansion and higher bandwidth) and sophisticated software (neural networks, probability models and rules analysis). -- Mike Covert, chief operating officer, Infinis Inc., Columbus, Ohio ... Automatic insurance decisions. By 2009, 50% of all insurance underwriting decisions will be automated using data mining technology. -- Richard Vlasimsky, chief technology officer, Valen Technologies."
>>> Business, Banking, Customer Relations, Image Understanding, Data Mining, Expert Systems, Neural Networks, Probability, Machine Learning, Reasoning, Vision, Systems, Applications

June 18, 2004: Breeding Race Cars to Win. By Michelle Delio. Wired News. "A technology that allows robots to rebuild themselves and computer programs to evolve and become better on their own is now being used to breed super-fast Formula One race cars. ... The breeding was done solely with computer-generated simulations using genetic algorithms -- programs that combine Mother Nature's laws and computer science to mimic the natural process of evolution. Using this sort of programmed procreation, the Digital Biology Interest Group [at University College London] has made self-healing battlefield surveillance robots -- gadgets that look like robotic snakes that can figure out how to wiggle home even when severely damaged, unlike less-evolved robots that typically just give up when one of their critical components goes out of commission."
>>> Genetic Algorithms, Engineering, Machine Learning, Robots, Applications

June 16, 2004: Research: From lab to market. By Michael Kanellos. CNET News. "Data mining, the ability to find unexpected patterns in accumulated data, was born during a lunch break. At a customer conference in the early 1990s, an executive at British department store chain Marks & Spencer was explaining his database woes to Rakesh Agrawal, an information retrieval specialist at IBM. The store was collecting all sorts of data but didn't know what to do with it. So Agrawal and his team began devising algorithms for asking open-ended queries, eventually authoring a 1993 paper that would become required reading in data-mining science. The report has been cited in more than 650 other studies, making it one of the most widely cited papers of its kind. ... Agrawal, the data-mining pioneer, is today working on a system that will scramble customer data in a way that will allow companies to study buying trends or other patterns while preserving strict privacy. ... In its Beijing labs, researchers are tinkering with handwriting recognition systems for Asian languages and a digital home in which appliances--lights, alarm systems, dishwashers, computers--can be operated through voice commands."
>>> Data Mining, History, Machine Learning, Natural Language Processing, Networks, Information Retrieval, Marketing, Smart Houses, Applications, Ethical & Social Implications

June 15, 2004: NASA Evolutionary Software Automatically Designs Antenna. Press release available from SpaceRef. "NASA artificial intelligence (AI) software - working on a network of personal computers - has designed a satellite antenna scheduled to orbit Earth in 2005. The antenna, able to fit into a one-inch space (2.5 by 2.5 centimeters), can receive commands and send data to Earth from the Space Technology 5 (ST5) satellites. ... NASA scientists have spent two years developing the evolutionary AI software that designed the antenna. 'The AI software examined millions of potential antenna designs before settling on a final one,' said project lead Jason Lohn, a scientist at NASA's Ames Research Center, located in California's Silicon Valley. 'Through a process patterned after Darwin's 'survival of the fittest,' the strongest designs survive and the less capable do not.' The software started with random antenna designs and through the evolutionary process, refined them. The computer system took about 10 hours to complete the initial antenna design process. ... 'Not only can the software work fast, but it can adapt existing designs quickly to meet changing mission requirements,' he said. ... Scientists also can use the evolutionary AI software to invent and create new structures, computer chips and even machines, according to Lohn. ... 'The software also may invent designs that no human designer would ever think of,' Lohn asserted."
>>> Genetic Algorithms, Engineering, Space Exploration, Telecommunications, Machine Learning, Applications

June 14, 2004 [issue date]: Innovators / Artificial Intelligence: Forging the Future - Rise of the Machines - These visionaries are making robots that can perform music, rescue disaster victims and even explore other planets on their own. By Dan Cray, Carolina A. Miranda, Wilson Rothman, Toko Sekiguchi. Time Magazine. "The Bionic Engineer - Driving School On Mars. Television critics will tell you that The Bionic Woman was just another cheesy '70s sci-fi series, but for Ayanna Howard it was a springboard to a career. When she was 12 years old, she became so captivated by the show's cyborg premise that she started reading books that reaffirmed the concept of integrating machines with humans. A thousand reruns and an electrical-engineering Ph.D. later, she's creating robots that think like humans for NASA's Jet Propulsion Laboratory. ... Three years ago, hoping to encourage others to follow in her footsteps, Howard launched a math-and-science mentoring program for at-risk junior high school girls. ... Howard hopes the program will help steer more young women into robotics, a field she says that within a decade will produce robots that mimic human thought processes. ... The Swarm Keeper - Metal Insects On Wheels. When James McLurkin was a high school junior on Long Island, N.Y., he built his first robot: a toy car that he rigged with a keypad, an LED display and a squirt gun. ... Now a graduate student in computer science at M.I.T., the young scientist is on the forefront of developing 'swarmbots'--packs of dozens of small robots that communicate with one another and work in harmony to complete an assignment. They have no centralized command system and can cover vast terrain; if one is destroyed, others fill in. ... Rescuer By Remote - Need Help? Send In The Robot. Within 24 hours of the 9/11 attacks on the World Trade Center, Robin Murphy was on the scene with a team of robots to help sort through the debris. It was the first real-world test of the Center for Robot-Assisted Search and Rescue in Tampa, Fla., the only unit of its kind on the planet. ... The Mimic Maker - The Android Who Learned To Dance. Mitsuo Kawato is fascinated with the brain -- so he helped build one. The biophysics engineer and computer researcher led a team at the Advanced Telecommunications Research Institute International in Kyoto, Japan, that spent five years constructing a humanoid equipped with artificial intelligence. Completed in 2001, the 6-ft. 2-in., 175-lb. robot was named Dynamic Brain, or DB for short. Says Kawato: 'We built an artificial brain hoping that it'll help us understand the real one.' ... So far, the robot has acquired about 30 skills, including juggling, air hockey, yo-yoing, folk dancing and playing the drum."
>>> AI Overview, Space Exploration, Neural Networks, Reasoning, Robots, Multi-Agent Systems, Artificial Life, Military, Hazards & Disasters, Applications, Machine Learning, Cognitive Science, Careers in AI (@ Resources for Students)

June 13, 2004: A Computer That Has an Eye for Van Gogh. By Douglas Heingartner. The New York Times (no fee reg. req'd.). "Who can say for sure that a great artwork is the real deal? ... Now a team of researchers at the University of Maastricht, here in the Netherlands, are taking a stab at rationalizing connoisseurship, a word that in its art-historical context refers to the formal process of determining who created a work of art. They have developed a computer system that quickly examines hundreds of paintings for telltale patterns. The results, they say, can lend credence to existing attributions or help dismiss them. Members of the team make modest claims for their system. 'The computer will come up with data that show some patterns, but we cannot decide whether these patterns are meaningful or not,' said Dr. Eric Postma, the leader of the project, known as Authentic, which is currently analyzing all paintings attributed to Vincent van Gogh. 'For that purpose we need experts. We can provide them with numbers, and they can interpret the numbers. And this interaction is where the real value of the project is.' ... Dr. Postma compares this pattern-seeking technique to chess. ... This is not the first time artificial intelligence has been used in authentication. In Germany in 1998, a team at the University of Bremen's Center for Computing Technologies trained their computer to identify the drawings of Delacroix, which it managed to do with 87 percent accuracy. ... In a more recent project at the Catholic University of Rio de Janeiro, a computer distinguished between 23 paintings made by the popular Brazilian painter Candido Portinari and five by his contemporary Enrico Bianco."
>>> Art, Pattern Recognition, Chess, Machine Learning, Law Enforcement, Applications

June 10, 2004: A golden vein - Computing: Analysis of customer information, better known as "data mining", is finally delivering on its promises-and expanding into some promising new areas. The Economist Technology Quarterly. "In the old days, knowing your customers was part and parcel of running a business, a natural consequence of living and working in a community. But for today's big firms, it is much more difficult: a big retailer such as Wal-Mart has no chance of knowing every single one of its customers. So the idea of gathering huge amounts of information and analysing it to pick out trends indicative of customers' wants and needs -- data mining -- has long been trumpeted as a way to return to the intimacy of a small-town general store. But for many years, data mining's claims were greatly exaggerated. ... In recent years, however, improvements in both hardware and software, and the rise of the world wide web, have enabled data mining to start delivering on its promises. Richard Neale of Business Objects, a software company based in San Jose, California, tells the story of a British supermarket that was about to discontinue a line of expensive French cheeses which were not selling well. But data mining showed that the few people who were buying the cheeses were among the supermarket's most profitable customers -- so it was worth keeping the cheeses to retain their custom. As data mining has matured, examples like this are plentiful. ... The traditional British pub seems like an unlikely place to find the latest in data mining. But some pub chains now change the prices of different drinks from day to day, using software that assesses the impact that 'happy hour' offers have on sales. ... Privacy advocates have long been wary of data mining, demonising supermarket loyalty cards, for example, as 'spies in your shopping'. Like any technology, of course, it can be misused. ... Forrester predicts that sales of BI [business intelligence] software, currently around $2 billion a year, will grow by 8.5% a year over the next three years. If new tricks like predictive analytics and unstructured-data analysis catch on, that could prove to be a conservative figure."
>>> Data Mining, Marketing, Business, Fraud Detection & Prevention, Law Enforcement, Machine Learning, Applications, Ethical & Social Implications, Industry Statistics

June 10, 2004: Fuzzy logic and neural nets: still viable after all these years? Though no longer headliners, fuzzy logic and neural networks are options in tackling challenging applications. By Graham Prophet. EDN Magazine. "[B]oth still have their place in your engineering tool kit. The two techniques are essentially unrelated, except that they both provide control methodologies to handle highly nonlinear or poorly specified problems, they both came to some prominence at about the same time, and they both faded from view in much the same way. Both neural networks and fuzzy logic aspire to allow electronic systems, built with familiar circuit techniques or employing conventional computing technologies, to attack certain problems in a way that mimics human responses and abilities. ... One of the intimidating aspects of fuzzy logic is the name itself, which has connotations of imprecision. On the contrary, however, fuzzy logic is capable of precise responses. It allows systems built around Boolean logic, handling binary values, to work with imprecisely defined values that you might express verbally as 'more,' 'less,' 'high,' 'low,' and so on. ... Neural networks, unlike fuzzy logic, seek to reproduce the versatility of the human brain in recognizing the end-to-end, input-to-output behavior of a system without understanding all the processes taking place within it. Taking as a fundamental model the interconnections of nervous systems within the brain -- neurons and synapses -- neural networks have the attributes of memory and learning. ... What happens to the expertise built up in neural and fuzzy techniques from their first flush of popularity? If you set about tracking down some of the pioneering companies from as much as a decade ago, you'd find that, although many no longer exist, some have transformed themselves into software-design and consultancy operations. These businesses are applying the same neural and fuzzy techniques but mainly in software simulation running on conventional computers, in areas such as financial modeling, financial services, and data mining."
>>> Fuzzy Logic, Neural Networks, Applications, Banking, Finance & Investing, Data Mining, Reasoning, Machine Learning

June 10, 2004: Brain learns like a robot - Scan shows how we form opinions. By Tanguy Chouard. Nature Science Update. "Researchers may have pinpointed the brain regions that help us work out good from bad. And their results suggest that humans and robots are more alike than we may care to admit, as both use similar strategies to make value judgements. ... The team also plotted brain activity on a graph to give a mathematical description of processes that underlie the formation of value judgements. The patterns they saw resembled those made by robots as they learn from experience. 'The results were astounding,' says study co-author Peter Dayan. 'There was an almost perfect match between the brain signals and the numerical functions used in machine learning,' he says. This suggests that our brains are following the laws of artificial intelligence."
>>> Cognitive Science, Machine Learning, Robots

June 8, 2004: Man who cracked computer engima. Opinion by Andrew Hodges. Edinburg Evening News / available from Scotsman.com News. "[Alan] Turing was fascinated by the concept of creating a mathematical machine to represent thought processes, and it was the 'Turing Machine' which became the foundation of the modern theories of computer science. He also envisaged a 'Universal Turing Machine' - one machine for all possible tasks - which embodied the essential principle of the computer. Turing's originality lay in seeing the relevance of mathematical logic to a problem originally seen as one of physics. He made a bridge between thought and action, which crossed conventional boundaries. All this was when he was just 24. Then he left Cambridge for a spell at Princeton and right away saw a link from 'useless' logic to practical purposes. ... In 1944, following the invasion of Normandy that Allied control of the Atlantic allowed, Alan Turing was almost uniquely in possession of three key ideas - his own 1936 concept of the universal machine, the potential speed and reliability of electronic technology and the inefficiency in designing different machines for different logical processes. Combined, these ideas provided the principle, the practical means and the motivation for the modern computer. ... From October 1947, the National Physical Laboratory allowed, or perhaps preferred, that he should spend the academic year at Cambridge. Out of this came a pioneering paper on what would now be called neural nets. ... Though marginalised in practice, he published his theoretical ideas on artificial intelligence in 1950 in a paper which is now one of the most quoted in science. His 'Turing Test' for intelligent machinery now has a long and entertaining history."
>>> History, Alan Turing (@ Namesakes), Neural Networks, Turing Test, Machine Learning

June 7, 2004: Cognitive Personal Assistant. AI-based systems could handle routine administrative tasks. Future Watch by Thomas Hoffman. Computerworld. "Researchers at Carnegie Mellon University are developing a computer-based administrative assistant that draws upon artificial intelligence (AI) techniques to perform routine tasks such as scheduling meetings for busy managers and filtering and prioritizing their e-mail. ... The project, called Radar (short for Reflective Agent with Distributed Adaptive Reasoning), is being funded by the Defense Advanced Research Projects Agency under a program called PAL, or Personalized Assistant that Learns. ... Using AI, Radar will draw on statistical and symbolic learning. Say a manager demonstrates a tendency to deny e-mail requests to hold meetings on Fridays over the course of a few months. Radar will pick up on this pattern and send a message to the manager asking whether the manager prefers to avoid meetings on Fridays."
>>> Agents, Natural Language Processing, Machine Learning, Probability, Applications

June 4, 2004: Programs of the Mind. Review by Gary Marcus. Science Magazine (subscription required). "Eric Baum's What Is Thought? [MIT Press, Cambridge, MA, 2004], consciously patterned after [Erwin] Schrödinger's book [What Is Life?], represents a computer scientist's look at the mind. Baum is an unrepentant physicalist. He announces from the outset that he believes that the mind can be understood as a computer program. Much as Schrödinger aimed to ground the understanding of life in well-understood principles of physics, Baum aims to ground the understanding of thought in well-understood principles of computation. In a book that is admirable as much for its candor as its ambition, Baum lays out much of what is special about the mind by taking readers on a guided tour of the successes and failures in the two fields closest to his own research: artificial intelligence and neural networks. ... Advocates of what the philosopher John Haugeland famously characterized as GOFAI (good old-fashioned artificial intelligence) create hand-crafted intricate models that are often powerful yet too brittle to be used in the real world. ... At the opposite extreme are researchers working within the field of neural networks, most of whom eschew built-in structure almost entirely and rely instead on statistical techniques that extract regularities from the world on the basis of massive experience."
>>> AI Overview, Cognitive Science, Philosophy, Neural Networks, Machine Learning

FOR MORE ARTICLES, SEE THE MACHINE LEARNING NEWS ARCHIVE