Dec. 26, 2014

After decades of start-and-stop, artificial intelligence is being advanced by major computing firms from Facebook and Google to IBM.

There was a burst of enthusiasm in the late 1950s and early 1960s that fizzled due to a lack of computing power, Then there was a great burst around 1985 and 1986 because computing power had gotten cheaper and people were able to do things they had been thinking about for a long time. The winter came in the late 1980s when the enthusiasm was followed by disappointment, and small successes did not turn into big successes. And since then, as soon as we get anything to work reliably, the industry stops calling it AI.
Today, thanks to the availability of vast amounts of online data and inexpensive computational power, especially in the cloud, we are not hitting the wall anymore," Hammond says. "AI has reached an inflection point. We now see it emerging from a substrate of research, data analytics and machine learning, all enabled by our ability to deal with large masses of data. (Read the main article)

Nov. 22, 2014

Artificial intelligence is picking up steam and is clearly on the mind of the biggest companies in tech. Of note, Facebook CEO Mark Zuckerberg recently called the development of artificial intelligence one of his top three key goals over the next 10 years. IBM launched a $100M Watson venture fund in January. And Google has already acquired a handful of AI startups including DeepMind Technologies of the UK for $500M.

CB Insights has picks a few up-and-coming startups that might be worth keeping an eye on in machine learning, natural language processing and who are working in the area of AI.

May 28, 2014

Google has released some new research about it efforts to maximize performance and minimize energy use at data centers through machine learning today. Long story short: Google is building superintelligent server farms that can learn from their past performance and improve themselves in the future.

Google’s AI data centers are a 20 percent project – the result of an employee, Jim Gao, working on something he found interesting that falls outside of his standard job description. Google is famous for allowing its employees 20 percent of their work time to come up with passion projects and things that they wouldn’t otherwise be able to work on. Thinking, learning data centers happened to be Gao’s main area of interest.

January 24, 2014

Adding to Google’s mixture of eccentric acquisitions is word that this week it has acquired artificial intelligence (AI) startup DeepMind, a London-based company the tech giant bought up for an estimated minimum of $500 million. According to Re/code, the purchase “is in large part an artificial intelligence talent acquisition.” Re/code notes that DeepMind has a team of at least 50 people and has secured more than $50 million in funding calling it “the last large independent company with a strong focus on artificial intelligence".

DeepMind joins a growing list of robotics and AI companies recently purchased by Google, including Boston Dynamics, its eighth acquisition of a Robotics Company in the past few months. The robots manufactured by Boston Dynamics possess locomotive abilities replacing the conventional wheel-based robots with ones that look and act more like humans or even certain kinds of animals.

December 14, 2013

Facebook users upload 350 million photos onto the social network every day, far beyond the ability of human beings to comprehensively look at, much less analyze. And so that’s one big reason the company just hired New York University (NYU) machine learning expert Yann LeCun, an eminent practitioner of an artificial intelligence (AI) technique known as “deep learning.” As director of Facebook’s new AI laboratory, LeCun will stay on at NYU part time, while working from a new Facebook facility on Astor Place in New York City.

September 26, 2013

Over the past decade, Microsoft co-founder Paul Allen has committed $500 million toward understanding the brain through his Allen Institute for Brain Science — aiming to ultimately transform the treatment of related diseases and disorders including autism, Alzheimer’s disease, and depression.

And earlier this month, he launched the Allen Institute for Artificial Intelligence — “AI-squared,” as he calls it — tapping longtime University of Washington computer science professor Oren Etzioni to lead a new quest for the elusive goal of computers that can acquire human levels of knowledge, reason and understanding.

During a symposium today at his EMP Museum, marking the Allen Institute’s 10th anniversary, Allen explained how the two initiatives relate to each another.
“In a way, it’s a strange kind of race,” he said. “Can you create an artificial object or entity … before you understand how it’s done in the brain? It’s a kind of crazy race. I don’t know which horse to bet on. I’m betting on both. Both are fascinating.”

April 25, 2013

Google has always been an artificial intelligence company, so it really shouldn’t have been a surprise that Ray Kurzweil, one of the leading scientists in the field, joined the search giant late last year. Nonetheless, the hiring raised some eyebrows, since Kurzweil is perhaps the most prominent proselytizer of “hard AI,” which argues that it is possible to create consciousness in an artificial being. Add to this Google’s revelation that it is using techniques of deep learning to produce an artificial brain, and a subsequent hiring of the godfather of computer neural nets Geoffrey Hinton, and it would seem that Google is becoming the most daring developer of AI, a fact that some may consider thrilling and others deeply unsettling. Or both.

Here is a snippet from his interview with Wired magazine:

WIRED: Are you participating in Jeff Dean’s program there to build an artificial “Google Brain?”

RAY KURZWEIL: Well, Jeff Dean is one of my collaborators. He’s a fellow research leader. We are going be using his systems and his techniques of deep learning. The reason I’m at Google is resources like that. Also the knowledge graph and very advanced syntactic parsing and a lot of advanced technologies that I really need for a project that really seeks to understand natural language. I can succeed at this much more readily at Google because of these technologies.

February 01, 2013

The European Commission has announced, the Human Brain Project, a $1.5-billion Supercomputer to simulate the Human Brain. (see the introduction video here)

In what is the largest and most significant effort to re-create the human brain to date, an international group of researchers has secured $1.5 billion to fund the incredibly ambitious Human Brain Project. For the next ten years, scientists from various disciplines will seek to understand and map the network of over a hundred billion neuronal connections that illicit emotions, volitional thought, and even consciousness itself. And to do so, the researchers will be using a progressively scaled-up multilayered simulation running on a supercomputer.
And indeed, the project organizers are not thinking small. The entire team will consist of over 200 individual researchers in 80 different institutions across the globe. They're even comparing it the Large Hadron Colllider in terms of scope and ambition, describing the Human Brain Project as "Cern for the brain." The project, which will be based in Lausanne, Switzerland, is an initiative of the European Commission.

January 04, 2013

This goes way beyond water wings. An AI system is learning to recognise the panicky movements people make when they are drowning. The idea is that the system could be used to save children's lives when there are no lifeguards around. Ultrasonic systems at swimming pools can alert lifeguards if someone is underwater too long - but few pools have them and they are no help at unguarded river banks and beaches. So Ken Sakamura and colleagues at the University of Tokyo created an AI system which could, for example, activate a body-worn flotation bag in an emergency.

The team want to make the system more robust by training it using more volunteers, with an accelerometer added to the sensor pack. They will present their results at the International Conference on Consumer Electronics in Las Vegas, Nevada, on 12 January.

October 22, 2012

Trending new startup, ITM Financial, who introduced the Social Sentiment Index into the forex market this year, has announced the release of their new software, Neural Network Forex Trading. Neural network forex trading uses the technology advancements discovered in recent years in the fields of DNA sequencing and brain mapping. The term neural network has traditionally been used to refer to a network or circuit of biological neurons. The modern usage of the term can sometimes refers to artificial neural networks, which are composed of artificial nodes. "By using the advances in science and analytics to our advantage in the financial markets, it has been a great discovery for our trading desk as well as our clients," says CEO, Curt Dalton.

September 09, 2012

Computers may soon be able to identify cities by just looking at random photos, if a new machine learning program by US and French succeeds. The machine learning program looks for details and characteristics that are unique to a city, tech site CNET reported.

Researchers at Carnegie Mellon University and INRIA/Ecole Normale Superieure in Paris fed 40,000 Google Street View images of Paris, London, New York, and Barcelona, as well as eight other cities to the system, for it to find "frequent and unique elements."

"Our data mining technique was able to go through millions of image patches automatically -- something that no human would be patient enough to do," it quoted Alexei Efros, CMU associate professor of robotics and computer science, as saying.

August 21, 2012

Vicarious, a startup trying to discover the rules that govern intelligence, has raised $15 million in a first round of funding from tech luminaries including Good Ventures, the fund created by Facebook Co-founder Dustn Moskowitz and Peter Thiel’s Founders Fund. The money isn’t to help commercialize its technology however, it’s basically R&D spending for a big tech undertaking.

July 25, 2012

A high school junior has created a computer brain that can diagnose breast cancer with 99 percent sensitivity.
Seventeen-year-old Brittany Wenger of Sarasota, Fla., wrote a breast cancer-diagnosing app based on an artificial neural network, basically a computer program whose structure is inspired by the way brain cells connect with one another. She won grand prize at the Google Science Fair for her invention in ceremony held in Palo Alto, Calif. last night (July 23).

Wenger wanted to get her computer brains to work on breast cancer because the least invasive diagnostic test for the disease, called fine needle aspirate, is also the least certain one. Often, if results aren't clear, patients need to undergo a second biopsy with a bigger needle or even surgery. Wenger wanted to boost the less-invasive test's success rates.

May 17, 2012

The two hundred people packed into a small screening room in Midtown Manhattan on a recent Tuesday night made quite a throng. Engineers, venture capitalists, and entrepreneurs sipped Sam Adams and nibbled bits from a fruit plate. They were there to learn about CrowdControl, a New York startup that is melding human workers with artificial intelligence to create the next paradigm for global labor: crowd computing.
The crowd filtered into the theater, and Kirill Shenykman, a venture capitalist who had recently led a $2 million investment in CrowdControl, took the stage. “What we are trying to do is to transform human labor into something that scales like software,” he explained. “We’re trying to take people and make them into bits.”

April 03, 2012

A new breakthrough in neural networking might just lead to truly intelligent computers. Dubbed a ‘super-Turing’ network, the new approach makes the neural networks so common to artificial intelligence research work very much like how our brains do.
The super-Turing neural networks are capable of learning and morphing, completely rearranging their design every time a new fact is learned. Which is huge. It means that this neural network learns an order of magnitude more effectively and faster than more traditional neural networks. And, unlike traditional neural networks, Siegelmann’s model thrives when exposed to constant stimulation.

March 21, 2012

A team of scientists from the University of Gothenburg and Chalmers has demonstrated that nanocellulose activates the creation of neural networks, paving the way to construct a 3D brain model to take brain research to an entirely new level.
The research work on growing human nerve cells on nanocellulose has been carried out for more than a period of two years. Professor Paul Gatenholm, one of the researchers from Chalmers, explained that until recently the researchers were not able to prevent the cells from dying after some time, because they failed to attach the cells to the scaffold. However, after a string of experiments, they developed a stable method to make the scaffold more positively charged, enabling the cells to get attached to the scaffold.

November 1, 2011

Tokyo: Robots that learn from experience and can solve novel problems, just like humans. Sound like science fiction, doesn’t it? But a Japanese researcher is working on practically realising this theory, with machines that can teach themselves to perform tasks they have not been programmed to do, using objects they have never seen before. The Self-Organising Incremental Neural Network, or “SOINN”, is an algorithm that allows robots to use their knowledge — what they already know — to infer how to complete tasks they have been told to do. SOINN examines the environment to gather the data it needs to organise the information it has been given into a coherent set of instructions.

September 12, 2011

Narrative Science, a startup based in Chicago, Illinois, is working to put writers like me out to pasture. Founded by Northwestern University researchers Kris Hammond, Larry Birnbaum and Stuart Frankel, the company offers software that takes data — sports statistics, financial reports, and so forth — and turns it into articles like this one.
For now, humans remain in the driver’s seat, guiding the software to make better selections as it writes more articles. The real litmus test will be when Narrative Science lets its software write its own press releases.

August 18, 2011

IBM has unveiled an experimental chip that borrows tricks from brains to power a cognitive computer, a machine able to learn from and adapt to its environment.
Reactions to the computer giant’s press release about SyNAPSE, short for Systems of Neuromorphic Adaptive Plastic Scalable Electronic, have ranged from conservative to zany. Some even claim it’s IBM’s attempt to recreate a cat brain from silicon.
“Each neuron in the brain is a processor and memory, and part of a social network, but that’s where the brain analogy ends. We’re not trying to simulate a brain,” said IBM spokeswoman Kelly Sims. “We’re looking to the brain to develop a system that can learn and make sense of environments on the fly.”

July 26, 2011

Researchers at Universitat Autonoma de Barcelona have developed an electronic tongue which can identify different types of cava wines, thanks to a combination of sensor systems and advanced mathematical procedures. The device automatically produces classifications similar to those of a sommelier. In order to design the electronic tongue, researchers from the UAB Group of Sensors and Biosensors, led by professor Manel del Valle, identified different cava samples using voltammetric measurements. Thanks to a combination of chemical measurement systems and advanced mathematical procedures - principal component analysis (PCA), discrete wavelet transform (DWT), and artificial neural network (ANN) - researchers achieved to copy the human taste system and distinguish between different types of cava, thus obtaining a classification similar to that of a sommelier. Through the use of the second order standard addition method (SOSAM) it was possible to quantify the amount of sugar added in the cava production process, demonstrating the efficiency of these processing tools.

May 25, 2011

Neural Technologies announced the launch of a software program designed to intelligently identify potential fraudulent claims made to state and federal health care programs. The real-time fraud detection program can identify unique markers and call attention to processors before a payment is made. Neural Technologies' solutions are based upon the culmination of over 20 years' extensive research and development into artificial intelligence and advanced neural processes. Using this comprehensive knowledge and experience, the company has developed a powerful, proprietary neural architecture that sets it apart from the competition.

December 14, 2010

Predictive maintenance improves profitability of Petrobras Zarate polystyrene manufacturing plant in Argentina by more than $1 million per year. But the benefits did not end there. A comprehensive asset management program implemented on approximately 1,000 intelligent FOUNDATION fieldbus and HART field devices has enabled continued productivity gains and greater profitability. Proactive maintenance measures and neural networks helped better utilization of raw materials, quick correction of flow deviations, and avoidance of low quality scrap. Plant availability soared to 99%, and annual production rose by 3 to 5%. Production reached an all-time high—66,000 tons of high-impact and crystal polystyrene annually.

November 27, 2010

The aim of the present study was to assess the relationship between lower urinary tract symptoms, anatomical findings, and baseline characteristics in women with pelvic organ prolapse (POP).

A cross-sectional observational study was performed, enrolling consecutive women seeking cares for lower urinary tract symptoms (LUTS) with evidence of POP. Data regarding baseline characteristics, LUTS, and physical examination were gathered for each patient. Multivariate analysis (multiple linear regression (MLR)) and artificial neural networks (ANNs) were performed to design predicting models.
LUTS result form a fine interaction between baseline characteristics and anatomical findings and demonstrates that ANNs are valuable instrument for better understanding complex biological models.

August 26, 2010

Health Discovery Corporation is a molecular diagnostics company that uses advanced mathematical techniques to analyze large amounts of data to uncover patterns that might otherwise be undetectable. It operates primarily in the emerging field of personalized medicine where such tools are critical to scientific discovery. Its primary business consists of licensing its intellectual property and developing its own product line of biomarker-based diagnostic tests that include human genes and genetic variations, as well as gene, protein, and metabolic expression differences and image analysis in digital pathology and radiology.

In medical applications, for instance, predictive models are built to make diagnoses. For instance, blood samples or tissue images may be examined to decide whether a patient is healthy or diseased. In the Active Learning Challenge, the participants were facing problems of document analysis, text classification, pharmacology, embryology, marketing, and ecology.

June 03, 2010

Neural networks built for image recognition are well-suited for "seeing" sound.

New work from students at the University of Hong Kong describes a novel use of neural networks, collections of artificial neurons or nodes that can be trained to accomplish a wide variety of tasks, previously used only in image recognition. The students used a convolutional network to "learn" features, such as tempo and harmony, from a database of songs that spread across 10 genres. The result was a set of trained neural networks that could correctly identify the genre of a song, which in computer science is considered a very hard problem, with greater than 87 percent accuracy. In March the group won an award for best paper at the International Multiconference of Engineers and Computer Scientists.

April 12, 2010

Developing intelligent artificial hands for hand amputees, neural devices to help people suffering from vertigo, dizziness and other vestibular disorders and the possibility to see how your brain responds while learning are a few examples of European research carried out in the area of future and emerging information and communication technologies (FET) that are being presented in the European Parliament in Strasbourg today. Twelve outstanding science projects funded under the European Commission's Future and Emerging Technologies programme will be showcased at the exhibition on "Science beyond Fiction: an Excursion into Future and Emerging Technologies". Europe is taking the lead in FET by proposing to invest around €500 million in exploratory research into high risk future Information and Communication Technologies (ICTs).

December 04, 2009

How about a computer system that recognizes when someone is trying to do something malicious with its code? That’s one goal of a consortium focusing on cyber research, with an ultimate focus on recognizing and preventing cyber threats or terrorism.

Northrop Grumman, the defense and intelligence community contractor, is funding the consortium in partnership with Purdue University’s Center for Education and Research in Information Assurance and Security, MIT’s Computer Science and Artificial Intelligence Laboratory, and Carnegie Mellon University’s Cybersecurity Education and Research Center.

November 17, 2009

A computer with the power of a human brain is not yet near. But this week researchers from IBM Corp. are reporting that they've simulated a cat's cerebral cortex, the thinking part of the brain, using a massive supercomputer. The computer has 147,456 processors (most modern PCs have just one or two processors) and 144 terabytes of main memory — 100,000 times as much as your computer has. The scientists had previously simulated 40 percent of a mouse's brain in 2006, a rat's full brain in 2007, and 1 percent of a human's cerebral cortex this year, using progressively bigger supercomputers.

The simulation, which runs 100 times slower than an actual cat's brain, is more about watching how thoughts are formed in the brain and how the roughly 1 billion neurons and 10 trillion synapses in a cat's brain work together.

The researchers created a program that told the supercomputer, which is in the Lawrence Livermore National Laboratory, to behave how a brain is believed to behave. The computer was shown images of corporate logos, including IBM's, and scientists watched as different parts of the simulated brain worked together to figure out what the image was. Dharmendra Modha, manager of cognitive computing for IBM Research and senior author of the paper, called it a "truly unprecedented scale of simulation." Researchers at Stanford University and Lawrence Berkeley National Laboratory were also part of the project.

October 14, 2009

Forrester analysts take closer look at the 15 technologies to watch. Those include areas from cloud-based services, to prediction analysis, neural networks and process-centric business intelligence.

September 23, 2009

Can computers win the Turing Test? Imagine a day when a machine will say, "Move over Turing! You can no longer consider machines to be less smart than humans! After all, we can think too. We do all the thinking and processing and you take all the credit, just because you are our creator! ". That would be an awkward and exciting situation. To be honest, there is a valid argument here in this imaginary conversation. As naive as it may sound for now, let me assure you that such a scenario is not far away. Applications are becoming more and more logic-oriented and increasingly intelligent. Read on ..

August 31, 2009

Computer scientists in Taiwan say they've developed a neural network computer program that can classify music based on its beat and tempo. Mao-Yuan Kao and Chang-Biau Yang of National Sun Yat-sen University in Kaohsiung and Shyue-Horng Shiau of Chang Jung Christian University in Tainan said their system could assist music archivists with an automated approach that assigns a genre to each tune.

July 22, 2009

A detailed, functional artificial human brain can be built within the next 10 years, a leading scientist has claimed.
Henry Markram, director of the Blue Brain Project, has already simulated elements of a rat brain.
He told the TED Global conference in Oxford that a synthetic human brain would be of particular use finding treatments for mental illnesses.

The Blue Brain project was launched in 2005 and aims to reverse engineer the mammalian brain from laboratory data.

June 2, 2009

Neural networks can discover patterns in data and successfully predict the future trend. A small Canadian company, Addaptron Software - the developer of decision support tools for stock investors and traders - has developed NNSTP-2, neural network computer tool, to help stock traders in predicting stock prices within 1-60 days. NNSTP-2 predicts future share prices using Fuzzy Neural Network (FNN). It operates automatically when creating the FNN, training it, and mapping to classify a new input vector. The input data transformed to characteristic matrices before training FNN. The forecasting is based on automatic scan of different inputs periods (historical price and volume data) to define accuracy of each one by back testing. Then the final forecast is built on weighted averaging of all forecasts. Each weight is proportional to the accuracy of a certain input period forecast.

May 8, 2009

A team of researchers from the University of Alcala (UAH) and the Complutense University in Madrid (UCM) have invented a new method for predicting the wind speed of wind farm aerogenerators. The system is based on combining the use of weather forecasting models and artificial neural networks and enables researchers to calculate the energy that wind farms will produce two days in advance.

April 30, 2009

Creating a virtual model of the human brain is one thing. I do it all the time, doodling little cerebrums while I talk on the phone. But getting your model to behave just like its flesh-and-blood counterpart? That’s a Frankenstein moment right there. Researchers with the Swiss-based Blue Brain Project have just created a virtual pack of neurons that acts just like the real thing, and hope to get an e-brain up and running.

February 6, 2009

The solution allows to momentarily detect potential fraud while providing financial institutions risk indicators that guide them to immediately decide if they will accept or decline the transaction. Visa says it is the first in the industry to launch such kind of security product.
Advance Authorization is based on Visa's state-of-the-art neural networks that have proved effective in detecting unusual spending patterns and monitoring for fraud in individual accounts. The upgrade made to this release is that it is updating in real-time from every transaction performed on the Visa network worldwide. As threats and activities change, the tool updates itself to look for new patterns. Authorization occurs immediately every time an authorization request passes through the Visa network. During a purchase the data of each cardholder is evaluated by the Advance Authorization. Then the system assigns a risk score to each authorization request, and if it detects and links together unusual patterns of event level behavior, it assigns a compromised account risk condition code as well.

January 1, 2009

KnitMesh Technologies, the Welsh knitted wire design and manufacturing company, has teamed up with Liverpool University to improve mesh design using neural networking techniques. The neural network model is an intelligent software that allows KnitMesh Technologies to predict the performance of knitted mesh components. According to KnitMesh, use of the software will reduce development costs and lead times, while ensuring optimal design during production of new parts. The two-year research programme will focus on development of the neural network model to advance the design of knitted mesh for a variety of specifications.

November 29, 2008

Brainware, a young company that makes document scanning and content management software, has partnered with Fujitsu to sell pre-integrated data capture solutions based on Fujitsu's scanning hardware and Brainware's software, the companies announced last week. Brainware was founded in 2006 following a management buyout of a subsidiary of SER Systems, a German developer of enterprise content management and call center software. Today, the company has its U.S. headquarters in Virginia and offices in several European countries. Brainware claims that more than $100 million have been invested in its core "neural network" classification and search technology, also called Brainware, which forms the basis for the company's two major product suites: IDC-Distiller, an automated data capture solution used for scanning tasks; and Globalbrain, a search and retrieval product.

October 23, 2008

DARPA (Defence Advanced Research Projects Agency) is sponsoring a new competition to develop "... a highly versatile and sensitive broad-spectrum device capable of detecting odorants under the challenge of real-world conditions". Obviously the military and security implications are immense, but so is the potential to develop further knowledge about the inner workings of one of the most primitive CNS pathways.
Evolved Machines, Inc. (Palo Alto, CA) has been selected as a prime contractor to engineer an artificial olfaction system incorporating "brain-like" neural pattern recognition. In their press release, DARPA labeled the project "RealNose", and the agency emphasizes that it will be "the first program to tackle the separation of multiple odorants in the presence of unknown backgrounds that characterize the detection problems presented in real-world settings."

September 12, 2008

Lifeline Biotechnologies, a developer of innovative medical technology, has filed the patent application for its new neural network diagnostic system. As previously announced, Lifeline has been working diligently with a prominent Asian university to further develop and enhance the company's First Warning System. This patent application is said to be the result of that extensive work. Louis Keith, emeritus professor of obstetrics and gynecology at the Feinberg School of Medicine at Northwestern University, said: "We can now determine which of the thermal aberrations the First Warning System detects is similar to those associated with known cancers. Using a combination of five neural network constructions, we have broken this illusive barrier.

August 12, 2008

When used with data mining tools and street investigations, predictive analytics can be among the biggest electronic enemies of the fraud rings that cost insurers tens of billions of dollars annually. Predictive analysis is a leap forward in antifraud technology. It injects artificial intelligence into fraud detection, allowing insurers to uncover suspicious claim patterns earlier in the claim cycle. Basically, it compacts the time-space continuum for investigations.

July 25, 2008


Toronto, Ontario, Jul. 25, 2008 - Intelligence Realm Inc. has recently completed a simulation of 100 billion neurons, the estimated size of the human brain. The simulation used distributed computing and involved over 4000 computers, 3000 volunteers, 10000 processors, 180 TB of data and lasted for a couple of months.

This was the first simulation that bypassed the 100 billion level and used database files to store the data. The simulation is one of the first steps in a long-term project that is aiming to build a large-scale artificial intelligence by reverse engineering the brain.

July 17, 2008


European researchers are developing a software that will give robots the power to learn when a person is sad, happy or angry. The Feelix Growing project is putting together simple robots that can detect different parameters—facial expressions, voice and proximity—to determine emotional states. The aim of the project is to develop a robot that can serve humans with special needs, such as the ill and the elderly. Using adaptable neural networks, the robot can learn the correct way to respond to people's emotions from experience.

For instance, if someone shows fear, the robot can learn to change its behavior to appear less threatening. If someone seems happy, the robot can make a mental (or, I guess, digital) note of what brought on that response. And if someone seems upset and lonely, the robot can give her a pat on the back, offer her a stiff drink and say "Elaine, you deserved someone better than that dickwad anyhow."

June 26, 2008


Nivis, developer and integrator of wireless network technologies, together with Cisco are debuting an embedded wireless IP mesh technology using a 6LoWPAN solution for integrated device management at the Cisco Live technology showcase in Orlando, Florida.

Nivis is presenting a new wireless network technology that allows disparate devices and sensors to communicate via a sophisticated neural network system using the 6LoWPAN(IPv6) protocol.

April 29, 2008


Appian Technology is the leading manufacturer and supplier of high performance, high accuracy Automatic Number/License Plate Recognition (ANPR/ALPR) systems. Appian's ANPR products are based on a proprietary neural network recognition engine called Talon. Neural network technology is superior to any template based Optical Character Recognition (OCR) ANPR system, offering significantly higher performance and accuracy, typically better than 97%.

Talon is a software based processor designed to be installed on to modern computers running the Windows operating system. Software holds BOF2.2 Web Services accreditation allowing all customers to continue to meet National ACPO ANPR Standards.

March 27, 2008


ROCKETinfo, Inc. (OTCBB:RKTI), a pioneer in news monitoring, analysis and search technology, today announced the release of a new version of its flagship online news search engine and portal, Rocketnews.com. Rocketinfo addresses the core challenge in the news search business: relevant news, provided in a timely manner. The just-launched Rocketnews.com aims to set a new standard amongst Internet news providers by answering the question: In this era of too much news, how do you find exactly what you need?

October 23, 2007


EasyNN-plus 8.0s can generate multilayer neural networks from imported text files, images or grids with minimal user intervention. The user can produce training, validating and querying files using the facilities in EasyNN-plus or using any editor, word processor or spreadsheet that supports text files. EasyNN-plus can learn from training data and can self validate while learning.

September 2, 2007


Exams mean a lot of work for examiners But in future, computers could help them reclaim their summer holidays. Professor Sargur Srihari's research team at the University at Buffalo, New York, is developing software to fully automate the essay-marking process."Trying to analyse children's handwriting is a completely unexplored domain," says Professor Srihari. Exam scripts are scanned into the computer, the software reads the handwriting and translates it into computer type, and then grades the response as an examiner would, Professor Srihari explains.

August 11, 2007

Legend Advisory Corp., subsidiary of Waddell & Reed, has been using artificial intelligence to invest mutual funds into various assets. Legend has applied this practice when managing retirement plans, endowments, foundations, institutions and individual assets. "We’re doing some phenomenal stuff," said Jim Leos, of the Legend Group and Legend’s regional vice president for Arizona. "We wanted to figure out what was the most proactive and scientific way to manage money that takes out the human emotions greed and fear. We’ve done that."

August 02, 2007

Legend Advisory Corp., a registered investment advisor, today announced that it has received an initial $5 million in faculty retirement funds from The University of Arizona, Arizona State University and Northern Arizona University, collectively the Tri-Universities, to be managed on Legend Advisory’s Strategic Asset Management platform, which utilizes groundbreaking artificial intelligence.
“A decade ago, many people thought artificial intelligence was science fiction, but it is now breaking frontiers in so many fields,” says Dr. Terry Riffe, director of the University Teaching Center and financial author from the University of Arizona. When applied to investments, says Riffe, “It’s a groundbreaking financial tool that can be invaluable.” Dr. Riffe was a catalyst in the effort at the Tri-Universities to offer Legend Advisory’s money management programs to faculty members and staff members in the University’s Optional Retirement Program. Riffe anticipates contributions to the program will quickly grow from the initial $5 million, as the staff becomes more familiar with the notable program.

July 21, 2007

Stephen Thaler is such a wretched musician that his wife won't even let him sing in the shower. And yet the computer scientist is releasing a CD of new music.
Thaler's computers at his Maryland Heights, Mo., company, Imagination Engines Inc., are intelligent and creative enough to teach robots to walk, help a car decide whether the object it is about to back over is a child or a toy, create substances harder than diamonds and design toothbrushes. They work in a variety of different industries. In their spare time, the Creativity Machines, as he calls his computer programs, make the ultimate in personalized music.

May 30, 2007

Israeli researchers Itay Baruchi and Eshel Ben-Jacob of Tel-Aviv University have demonstrated through experiment that it’s possible to store multiple rudimentary memories in an artificial culture of live neurons. This is a critical step towards cyborg-like integration of living material into memory chips. To create a new memory in the neurons, the researchers introduced minute amounts of a chemical stimulant into the culture at a selected location. The stimulant induced a second firing pattern, starting at that location. The new firing pattern in the culture along coexisted with the original pattern. Twenty-four hours later, they injected another round of stimulants at a new location, and a third firing pattern emerged. The three memory patterns persisted, without interfering with each other, for over forty hours.

April 17, 2007

For the first time anywhere, a researcher at the Hebrew University of Jerusalem has succeeded in observing in vivo the generation of neurons in the brain of a mammal.
Dr. Adi Mizrahi of the Department of Neurobiology at the Alexander Silberman Institute of Life Sciences at the Hebrew University, used mouse models to study how neurons, or nerve cells, develop from an undifferentiated cellular sphere into a rich and complex cell. This has great significance for the future of brain research, said Dr. Mizrahi, since "the structural and functional complexity of nerve cells remains one of the biggest mysteries of neuroscience, and we now have a model to study this complexity directly."

March 25, 2007

Robots have feelings, too -- or at least they will -- pending the completion of a pan-European research project being led by a group of British scientists. The Feelix Growing project aims to design and build a series of robots that can interact with humans on an emotional level, and actually adapt their behavior in response to emotional cues from their human counterparts. The robots used in the project are simple designs, including some "off-the-shelf" models. The complexity lies in the software, which will construct artificial neural networks to pick up on human emotions exhibited via facial expressions, voice intonation, gestures and other behaviors.

January 17, 2007

LMS is a leader in the application of advanced mathematical modeling and neural networks for medical use. The LMS CALM(TM) Decision Support Suite provides physicians, nursing staff, risk managers and hospital administrators with clinical information systems and risk management tools designed to improve outcomes and patient care for mothers and their infants during labor and delivery.

December 26, 2006

The implantation of microchip in milching animals to check the misuse of bank loans and tracking the movement of wild animals by radio collars are very common now a days. But humans, in future, are also to be tagged for tracking their movements by imbedding a microchip in the body using neural network

December 11, 2006

Fund houses, hedge funds and institutional investors have taken the application of quantitative models in investment decision making to a new high. Quant models use a variety of techniques, such as fuzzy logic, neural networks, genetic algorithms, Markov models, fractal methods, and clustering techniques. The investment techniques they use draw more from physics than from economics. Quant models use extensive back testing of past data to create their investment algorithms, raising the issue that the past may not accurately represent the future. Some of the early techniques that used simple technical rules based on past price behaviour have been accused of being exercises in 'torturing data until it confesses'.

June 26, 2006

An "emotionally aware" computer being developed by British and American scientists will be able to read an individual's thoughts by analyzing a combination of facial movements that represent underlying feelings. "The system we have developed allows a wide range of mental states to be identified just by pointing a video camera at someone," said Professor Peter Robinson of Cambridge University in England. The scientists, who are developing the technology in collaboration with researchers at the Massachusetts Institute of Technology (MIT) in the United States, also hope to get it to accept other inputs such as posture and gesture.

April 21, 2006

Ohio State University has scanned mixed gas-liquid-solid flow in 3D using a technique previously employed only for 2D scans. "Capacitive tomography has been practiced for many years, but not in 3D," Professor Liang-Shih Fan at the University told EW. Providing the array has a presence in three dimensions, that it is not just a 2D array, said Fan, 3D results can be extracted. A favoured arrangement is two rings of six electrodes around an 80mm pipe.
"The image construction is unique," said Fan. "Mathematicians like to use iterative techniques, but these do not give good images. We use a neural network-based technique. Image reconstruction is established by introducing a 3Dsensitivity matrix."

March 20, 2006

"We are taking knowledge from neuroscience and using it to build better computers," said Kwabena Boahen, an associate professor in the Department of Bioengineering who directs a research group tasked with mimicking the functions of the brain's complex neural system using silicon chips. Boahen hopes his research will lead to small computers that could replace damaged neural tissue or silicon retinas that restore vision. He believes understanding how the brain functions could help make computation more efficient. "Soon after I got to the United States, I learned about neural networks and I thought [they were] really elegant," Boahen said. "You could present a bunch of examples to the network and it would learn." Boahen began studying very large scale integration, or VLSI, circuits. These experiences led him to design an associative memory chip that, using pattern recognition, would "learn" to associate pictures with the words used to describe them.

March 13, 2006

A neural network model can predict profitability based on customer age even though the relationship between the two variables is non-linear. Neural network models, Schumacher says, are most useful when the target variable has a high useful to irreverent ratio, or when interpretation is not the goal.
A neural network is a non-linear prediction. So a 20-year-old might be worth $1 in profit per year, a 30-year-old $4 in profit per year and a 40-year-old worth $5 per year. Since the profit prediction does not go up by a constant amount, and might even start to go down again for 60-year-old customers, you have a non-linear model, and that's what neural network models are designed for.

February 16, 2006

Conventional computing methods can solve most data processing and control tasks as long as you throw enough high-speed silicon at the problem.
Our brains, though, can complete some remarkably complex tasks, faster than a room full of computers, and yet we achieve this with neurons that do not respond in much less than a millisecond. That networks of biological neurons can often operate more efficiently than nanosecond-switching logic-gates is not startling news but applying that knowledge to building alternative models of computation has had mixed results. Anyone working in electronics a decade ago will remember the excitement, followed by disappointment, generated by fuzzy logic microcontrollers that used artificial neural network algorithms and machine-learning to ‘revolutionise’ embedded systems. There was no revolution. But the idea has not disappeared and today, driven by increasingly stringent emissions regulations, software and hardware-based neural network-based techniques are being successfully applied to engine control and diagnostics in automotive embedded systems.

February 02, 2006

An advanced controller is showing "promising results" by learning on-the-fly how to operate an engine cleaner and more efficiently, say researchers at the University of Missouri-Rolla. The researchers created a neural network controller that is implemented as a software program. Artificial neural networks are adaptive systems, which "learn" based on the successful connections they make between neurons or nodes. "The neural network observer part of the controller will assess the total air and fuel in a given cylinder in a given time," Sarangapani says. "It then sends that estimate to another neural network, which generates the fuel commands and tells the engine how much fuel to change each cycle."

November 15, 2005

Quintura Inc. the next-generation web search company, today announced the release of its revolutionary web search software, Quintura Search 1.0 beta that helps a user find the relevant information on the Web easier and faster. Today’s standard keyword search on the Internet offered by Google, Yahoo! and MSN returns thousands and millions of results. It is often not an easy task for a regular web user to build a more specific query, to narrow the search and find the relevant information. Quintura Search helps to overcome those limitations by offering a visual semantic map, the map of keywords and relationships between keywords. Adding or subtracting keywords from a query using the map and a mouse click, “One-Click Search”, allows a user to specify the context or meaning of the keyword, therefore narrowing the search and finding the relevant information faster.
The Quintura technology is based on over a decade of the founders’ innovative research and development in the area of neural network and artificial intelligence.

October 7, 2005

DARPA's robot racing challenge will pit artificially intelligent robots designed to drive autonomously against a hazardous, 150-mile desert course. The robot racers must balance care with speed and finish the course in less than 10 hours--and the odds are stacked against them.
Thrun (director of Stanford University's artificial-intelligence laboratory) and a team of computer scientists wrote more than 100,000 lines of code to tell it what to do. A map tells the car where to drive; a planning tool points out unsafe terrain; and a controller translates all of that into action. The software runs on six Pentium M processors, Intel-made, low-power chips originally designed for the telecommunications industry.

September 2, 2005

IRIS International, Inc. (NASDAQ: IRIS), a manufacturer and marketer of automated IVD urinalysis systems and medical devices used in hospitals and reference clinical laboratories worldwide, today announced that President and Chief Executive Officer Cesar Garcia will be presenting at ThinkEquity Partners LLC's 3rd Annual Growth Conference at 9 a.m. local time on Tuesday, September 13, at the Ritz Carlton Hotel, 600 Stockton Street, in San Francisco.
IRIS International, Inc. (www.proiris.com) is a leader in automated urinalysis technology with systems in major medical institutions throughout the world. The Company's newest generation iQ(R)200 Automated Urine Microscopy Analyzer, utilizing image flow cytometry, patented Automated Intelligent Microscopy (AIM) technology and neural network-based particle recognition, achieves a significant reduction in the cost and time-consuming steps involved in manual microscopic analysis. The Company's StatSpin(R) subsidiary, based in Norwood, Mass., manufactures innovative centrifuges and blood analysis products. Advanced Digital Imaging Research, LLC (ADIR), based near Houston, Texas, is the Company's imaging research and development subsidiary.

September 1, 2005

Intelligent computer software capable of predicting when systems are about to break down or need special attention is expected to improve operations and generate large cost savings. The technology has already been used to improve the reliability of high-power advanced chemical lasers, and nearby computer chip manufacturers are expected to save millions of dollars a year by installing the technology on just one portion of a production line.
"There are significant advantages to performing maintenance on high-value equipment when needed instead of on a periodic basis," noted Victor Stone, a computer engineer at the laboratory's Directed Energy Directorate. "The equipment can be safely operated longer, which improves productivity and saves money."
The technology, called Prognosis, uses advanced software to predict conditions, circumstances and faults. It is being developed by Mr. Stone and Dr. Mo Jamshidi, a professor at the nearby University of New Mexico and director of the university's Autonomous Control Engineering Center. Dr. Jamshidi is temporarily employed by the directorate under a special arrangement.

August 20, 2005

Using neural network software, British scientists want to establish a Digital Automated Identification System (Daisy) for all forms of life on Earth.
British scientists have unveiled plans to create a digital library of all life on Earth. They say that the Digital Automated Identification System (Daisy), which harnesses the latest advances in artificial intelligence and computer vision, will have an enormous impact on research into biodiversity and evolution.
Daisy will also give amateur naturalists unprecedented access to the world's taxonomic expertise: Send Daisy a camera-phone picture of a plant or animal and, within seconds, you will get detailed information about what you are looking at.

July 07, 2005

One of them for the Neural Matrix CCD: Initially designed to help scientists learn how neurons in the human nervous system communicate with each other, the Neural Matrix CCD is the first step in creating combined biological and electronic chip implants that can provide neural networks of living, interconnected nerve cells for testing drugs and sensing toxins for homeland security -- and, someday, restoring the use of limbs and eyesight and improved mental functions in patients.  In 2004, a team of scientists and engineers led by Eleanor Blakely and Ian Brown, including Kathy Bjornstad, Jim Galvin, Othon Monteiro, and Chris Rosen, developed a technique for growing the first large arrays of networked neurons on the prepared optical surface of a charge-coupled device (CCD). Diamond-like carbon deposited on the optical surface of the CCD is patterned in fine detail, then coated by a continuous layer of cell-culture collagen, and finally seeded with neurons. The coated CCDs now have millions of individual sensors that can record changes in electrical potential from individual nerve cells in real time while precisely mapping each neuron's activity within the neural network.

June 22, 2005
MSN Search Updates Results Based on RankNet. Besides the news yesterday that MSN Local has launched, the people at MSN Web Search snuck in an update to their search results with an algorithm based on what MSN calls RankNet. The search results seem more relevant to the query and MSN feels that RankNet “has imporved [their] relevance and most importantly gives [them] a platform they can move forward on.” The new ranking technology is based on neural net, which was discussed by Microsoft in a research paper headed by Chris Burges titled Learning to Rank using Gradient Descent.

June 17, 2005
The patent-pending solution is designed to detect potential fraud happening not just on individual cardholder accounts but throughout the whole Visa network in real time. It works by using neural networks to detect unusual spending patterns. When a card is swiped it sends the card issuer an instant rating of a transaction's potential for fraud. The issuer can then send an immediate response back to the merchant to accept or decline the transactions.

June 04, 2005
The research team in the IST project ERMIS, which focused on linguistic and paralinguistic cues in human speech and finished at the end of December 2004, created a prototype able to analyse and respond to user input. The team included researchers with skills ranging from engineering and computer science to psychology and human communication. In the analysis phase, the team extracted some 400 features of common speech, then selected around 20-25 as the most important in expressing emotion. These terms were then fed into a neural network architecture that combined all the different speech, paralinguistic and facial communications features. For facial expression, some 19 were selected as the most relevant and were input accordingly.