Dr. Gardiner Tucker, Director of Research (1963-67)

By Gardiner Tucker

I said about IBM’s research organization, when I joined the Watson lab at Columbia University in 1952, that it provided a wonderful degree of academic freedom, even though it wasn’t technically academic. That was the same spirit in which we started the Fellows program when I became director in 1963.

IBM Research had by the 1960s established itself at the forefront of a number of technical disciplines that we judged had the potential to lead to new hardware and software, as well as entire new fields of information systems. Recognizing our people for leading these breakthroughs was, at the time, through promotion to team leader or department manager.

What we needed was a way to encourage and reward individuals in a way that let them continue creative research, unencumbered by administrative duties. We also wanted to cultivate a way to encourage individual “gadflies” or “catalysts,” who could stimulate ideas in others, and help colleagues overcome bottlenecks.

This is why we decided to start the IBM Fellow program. We chose the name “fellow” by analogy with how universities recognized outstanding scholars.

Phyllis Baxendale and F.E. Firth studying the behavior of language and the organization of information from the point of view of machine manipulation.

One Research colleague who inspired me, and exemplified who a Fellow should be, was never actually named a Fellow. Phyllis Baxendale, who I knew from managing the research team in San Jose in the early 1960s, was programming a computer to do automatic abstracting, indexing and retrieval of documents. Her machine did this by extracting patterns of word usage from a set of documents, and comparing them with patterns from broader literature. This was pioneering work; an early harbinger, perhaps, of Watson?

Nathaniel Rochester is one of my favorite examples of a Fellow. Nat was already well known for designing the IBM 701 (the first general purpose, mass produced computer) and the first symbolic assembler. His CONCEPTOR project in 1955 attempted to simulate the brain’s neural network on an IBM 704 to more-efficiently detect patterns in a stream of data. Though it was not successful, we consulted John von Neumann about the architecture’s possibilities. He confirmed that it would have succeeded on a machine capable of simulating vastly more neurons than the 1,000 maximum on the 704.

Incidentally, at an annual meeting of the IBM board of advisors, I described the CONCEPTOR as part of my Research review. A shareholder reacted with “we’re in the business of business machines, not brain machines!” While the project was eventually dropped, it was this ability to think ahead of his time that earned Nat status as a Fellow in 1967.

There were outstanding individuals in many areas of Research. I picked these two because, while other industrial laboratories were also pursuing issues in device technologies and new electronics, IBM was unique in its equal emphasis on organization, programming and application methodologies – that is to say, on the revolution in information processing.

I am delighted to see that the Fellow program encompasses the entire company, and is still such an important part of IBM. When you hear the term “IBM Fellow,” think of a person who embodies a place with pioneering vision in an ever expanding field.

Note: Dr. Gardiner Tucker served as principal director of defense research and engineering in the U.S. Department of Defense, was named assistant secretary of defense in 1969, and assistant secretary general of NATO for defense support in 1973.

Technorati Tags: Big Data, CONCEPTOR, IBM 701, IBM Fellow, IBM Fellows Program, IBM Research, IBM Smarter Planet, ibmfellows, innovation

Update:

Teams of three students from MIT/Sloan and HBS take on IBM’s Watson. (This is only the second contest matching Watson against collegians. In the previous contest, Watson beat teams from Carnegie Mellon University and the University of Pittsburgh. Pitt came in second, much to the chagrin of rival CMU!)

Harvard wins the first question, with “What is Belize?” Answering: countries in central America, ending with “e”

But then Watson takes over, running the category.

The machine picks “Who’s Your Daddy Company?” as the next category, eliciting a huge hook of laughter from the audience.

They finished the Jeopardy! round, with Watson, $8600; Harvard, $5200  ; and MIT,  $-200 .

(I got disconnected from HBS’s Wi-Fi at a crucial moment, destroying the coverage of the second round. Grrrrr)

Final Jeopardy!

Clue: Finding the spot for this memorial caused its creator to say “Americans will march across that skyline.”

The question: Mt. Rushmore.

Harvard and Watson answer correctly. MIT does not.

Final score: Watson, $53,601; Harvard, $42,399; MIT, $100.

!!!!!

—————-

The Start:

9:45 a.m.

Erik Brynjolfson, MIT Sloan School of Management, kicks off by talking about a concern these days about technology and its role in society. Some people are saying that innovation has been stagnating, and that’s contributing to the slowness of the economic recovery.

“The issue isn’t that technology is stagnating, but that we haven’t been keeping up with technology. Societies, institutions and structures haven’t advanced rapidly enough to keep up with the advances. We’re creating a lot of wealth through technology, but the benefits aren’t going to regular people in the middle of the income distribution.”

This has been a great decade for productivity growth, even better the 1990s. It has contributed to income growth per capita. Yet median income growth has not improved much. “A lot of wealth has been created that goes to the people at the very top of the income brackets.”

Click here to view the embedded video.

———-

The context: Here’s the new book by Erik and Andrew McAfee, Race Against the Machine.

—————-

10 a.m.

IBM Fellow David Ferrucci talks about the making of Watson, IBM’s question-and-answer machine, which in February beat the top past champions on the TV game show Jeopardy! (He’s speaking again this afternoon, so I’m going to go into detail on that.) For now, here’s another IBM Fellow, Bernard Meyerson, talking about the importance of Watson.

Click here to view the embedded video.

——————

10:40 a.m.

Question to IBM’s David Ferrucci about Watson: How long until Watson can program itself?

Ferrucci:

It already does that, but will do so more in the future.

“I can imagine a situation where you’re searching for different models, different weights to answers, and it automatically programs itself to do that.”

He wouldn’t predict when.

——————–

10:45 a.m. Panel: What Can Technology Do Today and in 2020?

Andrew McAfee of MIT asks the question: Why are we seeing these astonishing advances now?

Afred Spector of Google: The Web makes it possible to combine a lot of information and access it via the Web. We also have a huge amount of feedback from users. And we have a large amount of software components. We can combine things and piece things together. “We’re solving a collection of problems which are acceptably probabilistic.”

Rod Brooks, an AI and robotics expert at MIT and entrepreneur: We have enough computer power to solve bigger and more complex problems. “Using machine learning and statistics we’ve managed to come up with algorithms which learn things acceptably well.”

David Ferrucci of IBM: “What’s exciting is the ability to generation hypothesis using induction and then track them back and evaluate evidence during an inductive process.”

———————

11 a.m. Panel: What Can Technology Do Today and in 2020?

Andrew McAfee of MIT asks the question: Why can’t computers do things that a two year old child can do?

Rod Brooks, AI and robotics entrepreneur “There is progress but it’s in narrow subfields. But it can do great things. Google cars are an example: They don’t do a lot of things but they do a few things very well.”

Afred Spector of Google:  Google Translate is another example. We’re up to 69 languages. We’re working on quality. “We want to get to the languages that are less spoken so all those populations will have access to the Web. We want to make the knowledge available to everybody.”

Another project: Making it possible so the machine automatically understands things so well that we can translate an image into a text description, or visa versa.

Brooks: He talks about the problem with manufacturing in the US. We keep going to high tech manufacturing, but that makes us too narrow. Not enough jobs created. He says we need to develop manufacturing that can employ a lot of people, by automating the low value pieces more but produce a wide variety of products.  “The answer is in the masses–creating robots that people can interact with and use.”

David Ferrucci of IBM: “I see a future where computers can act as intelligent mediators that enable informed collaboration, for instance, between you and your healthcare team, so you can make better decisions about your treatment.”

—–

Here’s McAfee talking about the coming capabilities of machines and their impact on jobs and job creation:

Click here to view the embedded video.

——————-

11:20  a.m.

Panel: What Can Technology Do Today and in 2020?

Andrew McAfee of MIT asks the question:  What would accelerate your work the most: data, computing power, or smart researchers?

Rod Brooks, MIT professor, and AI and robotics entrepreneur : We have enough computing power and data. “I have a bunch of smart PhDs, but you have to direct them in the right direction. You need the right reward structure for research.”

“We’d be better off if universities were smaller, had fewer people working form them, and focused on deep fundamental research. Let organizations like IBM do the applied research.”

Afred Spector of Google: “When we go to universities we’re surprised and disappointed that the faculties aren’t doing more high-risk research.”

Google sees a need for vastly more computing power. We need to do “deep learning.” It’s a new level of machine learning. Google has 5,000 PhD’s in computer science. “We need even more talent.”

“We need all three to get better.”

David Ferrucci of IBM:  “Researchers, data and machines, in that order.”

————

12:20 p.m. Panel: How Will Technology Affect Productivity and Employment?

Erik Brynjolfson, MIT Sloan School of Management, asks: What does technology mean for technology and jobs?

David Autor, economics professor at MIT:  There’s a long running debate. Does technology eliminate jobs? The stock answer is to call people who ask it Luddites.

We’ve seen incredibly rapid technology change over the past century and it eliminated a lot of farm jobs, but it created jobs elsewhere. “We’ve seen rising employment rates; and it raises productivity and incomes.”

However, there’s another side to this. Technology increases our efficiency but it can compete with workers and their skills. “Technology changes much faster than people can adapt.”

Middle-education and middle-skilled jobs are the vulnerable ones. That’s manufacturing jobs and administrative jobs.

“This creates real challenges. We should be worried. The set of opportunities are far more bifurcated than then used to be.”

“It’s leading to even more unequal distribution of wealth.”

Irving Wladawsky-Berger, former IBM executive and MIT lecturer:  He says he has been focusing on technology based innovation in the service economy. So many of the new jobs are in services. About 80% of the service jobs are information-based jobs. Technology will be used more and more in this area. So these jobs will increasingly come under pressure, too.

This is another period of creative destruction.

New industries will be created that will create the mid-skilled and mid-education jobs. “I don’t know the answer”

“The top-down approaches to job creation aren’t working. We have to rely more on bottoms-up approaches—entrepreneurialism.”

Frank Levy, a labor economist at MIT: Keeps things in perspective. Everything we see today is colored by the recession. It doesn’t have a lot to do with technology—but with the collapse of the housing bubble.

At the same time, the middle-skill job problem is very real.

“Because of the recession, it’s going to be hard to get kids to get advanced education at the same time that the jobs that will come will require advanced educations.”

———

12:45 p.m. Panel: How Will Technology Affect Productivity and Employment?

Erik Brynjolfson, MIT Sloan School of Management, asks: How do we create new jobs for mid-skilled people?

Irving Wladawsky-Berger, former IBM executive and MIT lecturer: Cloud computing and other technologies can help entrepreneurs get started and build companies and hire people. So a lot of small companies will spring up—not the high tech companies but companies that take advantage of technology.

David Autor, economics professor at MIT: That’s good, but it won’t produce a lot of jobs. “Most people want to be employed. They want to work for somebody. If they have a choice, that’s what they do.”

Frank Levy, MIT: He calls for apprenticeships and case-based education to bring up the skills.

Wladawky-Berger: Germany has done a better job at creating the mid-skilled jobs.

Autor: The Germans have adapted more quickly than other developed economies. They brought up the skills and reduced wages for mid-skilled people, which made the people and the country more competitive.

There are decent middle-skilled job in health care, repairs, the trades. But all require post-high schools investments in skills. “But you can’t go to the Harvard of plumbing.”

Wladawsky-Berger: Can’t community colleges help fill the void.

Autor: “They’d like to, but their funding is being cut by states and communities.”

—————————-

1 p.m. Panel: How Will Technology Affect Productivity and Employment?

Question from the audience: “I’m worried about how we communicate about the new capabilities of machines and their impact on jobs. Will people react against it?”

Irving Wladawsky-Berger, former IBM executive and MIT lecturer: There’s a consensus that just as we transitioned from the agricultural age to the industrial age, and literacy and education went up, in today’s world you need the next level of education. You need information-based literacy, and teamwork literacy. People who learn to use these tools can make a good living. If we can communicate that we’ll be okay.

Frank Levy, MIT: We should be clear about what machines can do and what they can’t do, and not talk a lot about the “singularity”—the point in the future when machines will be able to truly think.

“Just tell it straight in terms of what we know now. Don’t try to scare people.”

Here’s Wladawsky-Berger talking about the future of job creation:

Click here to view the embedded video.

——-

1:25 p.m. Remarks from Martin Fleming, chief economist at IBM:

There have been five waves of technological change over the past three centuries. With each wave, the new technologies fundamentally altered the way business—and work—was done. “The business changes because the technology makes it possible to do so.”

With IBM’s Watson, for example, you enable the democratization of clinical decision making. The practice of healthcare can be fundamentally changed. Evidence-based medicine is made possible.

Each wave was also accompanied by an economic crash, typically at the time when the new technology is impacting the old ways of doing things but has not yet produced all of the productivity gains that are coming on a mass scale.

We’re now entering into a period where the economy is beginning to open up opportunities for the deployment of significant new innovations. “Radical new technologies will be deployed. New industries will be created.”

“This is about the transformation of the economy.”

——

At Harvard Business School now…

3:30 p.m. David Ferrucci, head of IBM’s Watson project, talks about how Watson came to  be and where the technology is going.

Watson was a grand challenge aimed at driving important scientific advances. It gets people to think about the implications of technology–where is it today and where might it go.

Watson beat former grand champions at TV’s Jeopardy! quiz show.

He points out how much more difficult it is for a computer to have a conversation with a person than it is to play chess–a previous grand challenge that IBM took on in the 1990s when one of its machines beat the best chess player in the world. That’s because, in conversation, “context defines meaning.”

It’s a very difficult problem. Computers can’t relate words to experience.

“Jeopardy! helped us push the kind of technology that interprets natural language to determine meaning.”

————————-

4:00 p.m. David Ferrucci, the father of Watson, talks about how the technology came about and where it’s going.

We did 8000 experiments to develop Watson’s capability. There were lots of growing pains. Example: New York Times Headlines: An exclamation point was warranted for the “end of” this” in 1918. Watson’s answer: “a sentence.”

The project took 4 years.

Originally, running on a single PC, it took two hours for Watson to answer a single question. “The producers insisted that that would make for a boring game.” So they scaled the machine up to a 2880-core computing system.

The important thing about Watson is that it collects evidence and builds confidence in an ansers. when we think about applying it to medicine or law, we don’t think it’s providing the answer but providing useful suggestions–based on an evidence profile.

We’ll give a human decision maker the top answers to a question and the evidence and analysis that led Watson to those answers. This is about empowering the decision maker.

——————

Here’s Ferrucci talking about how the software program can help transform the healthcare industry.

Click here to view the embedded video.

Technorati Tags: Harvard Business School, HBS, IBM, MIT

This quest would be plenty challenging. Computer Scientists have been developing theories about quantum computing ever since physicist Richard Feynman first proposed the concept of computing based on quantum mechanical phenomena in 1982. Nearly 30 years later, there are no quantum computers.

Another proposition came in a close second, with 277 votes: “fight global warming.” (This one got my vote.)

Other suggestions ranged from the earnest, such as “take healthcare to the next level,” with 18 votes; to the ridiculous, “time travel,” with 97 votes.

We’ll pass along the top suggestions to the folks at IBM Research.

To read what it’s all about, see two previous posts, this one by IBM researcher Dario Gil about the effort to create learning systems, and this one, the live blogging stream from IBM Research’s colloquium, the Frontiers of IT.

Technorati Tags: IBM

http://www.e-masary.com/

E-Masary’s vision is to establish a REAL-TIME payment gateway for the masses that enriches their lives through job creation while jointly mobilizing and linking the lives of the banked with the un-banked.

Watch their 1-minute pitch, then VOTE for them by clicking “LIKE” below:

Click here to view the embedded video.

<< Previous Startup | Next Startup >> | Back to Main Menu

We asked students and faculty members at the ACM International Collegiate Programming Contest what they think Watson should do next, now that it has vanquished the former champions of Jeopardy! Here’s what they had to say.

Today’s computers are “smarter” than those of the mid-20th century, when Watson was busy trying to calm people’s fears about the onslaught of electronic brains.  But even with the achievements and potential of Watson, the Jeopardy-playing computer,  IBM is pushing the borders of machine capability with the goal of augmenting human thought, not replacing human brains with machines.

Still it’s intriguing to compare the two, and for scientists, nature offers models for next-generation computer design. At the conference IBM Research Colloquia – Zurich last week, Prof. Karlheinz Meier of  Heidelberg University spoke about neuromorphic computers during his presentation on Brain Inspired Computing. A key aspect of neuromorphic computing is understanding how the brain works and attempting to mimic some of its functions in silicon.

For me, the most intriguing moment during Meier’s presentation was when an audience member asked him a question that got to the heart of the man vs. machine debate. Essentially, he asked: Why copy nature? Why not aim to do better?

Question: “This may be a philosophical question. If you look at the properties of neurons and synapses and so on, do you see these wonderful devices or would it be possible to conceive that the brain evolved out of very simple cells, which represented a limited technology, and this was the best nature could do with what was available. If so, is it possible to invent other sorts of devices which could produce a better brain?

Meier: “Absolutely. That is a very valid question. We are today trying to copy what evolution has produced. It may not be the best thing. There may be more efficient solutions. That’s something we have to work with.”

One newsy element of the Zurich colloquium was the official launch of of a new $90 million Nanotechnology Center at IBM’s research laboratory in Zurich. The center is a joint venture of IBM and ETH Zurich, a premier European science and technology university. Meier said that the center, with it’s large collection of state-of-the-art nanotechnology tools, will be a valuable resource for scientists like him. “With this kind of facility and the tools and a rapid design cycle, we can try out all of these things,” he said.

In that building, where physics, biology, chemistry, and material science meet, the question of whether or not machines can be made that think may finally be resolved.

Technorati Tags: IBM, Karlheinz Meier, nanotechnology, neuromorphic computers

The company’s researchers and business leaders have been busy searching commercial uses for the technology, and they’re making progress. David Ferrucci, the IBM Fellow who heads up the Watson project, today told a group of journalists and analysts at a briefing on Big Data at IBM Research that the healthcare field is especially promising. IBM is developing applications in collaboration with physicians and researchers at Columbia University and the University of Maryland. Meanwhile, it took one researcher just three months to adapt the Watson Jeopardy! database to the medical field. Presumably, adaptations to other domains will be relatively easy, as well.

But a big issue is affordability. Watson is “embarrassingly parallel,” in computer science parlance—meaning the machine uses thousands of high-performance microprocessors. Embarrassing parallelism is expensive.

During a break form the briefing, I asked Ferrucci and other IBM colleagues about the affordability issue. Ferrucci’s response was that the cost of computing is dropping rapidly, and that the more industries Watson can serve and the more applications that are running on the platform, the more affordable it will be for each individual client.

Rod Smith, another IBM Fellow who heads up the Emerging Internet Technologies group at IBM Research, told me that another key will be the simplicity of the user interface. If IBM can develop user-friendly interfaces, clients will get rapid adoption of the technology and rapid returns on their investments. So, even if the service is relatively expensive, it will be worth the price.

Another key to affordability will be setting up Watson services for specific industries as shared services offered from the computing cloud. That way, many clients can use the same application and the same computing resources, making the services highly efficient.

It has been just three months since the Jeopardy! contest, but, already, it’s clear that the Watson machine has a life after Jeopardy! In the months ahead we’ll find out whether it can play a major role in making the world work better.

Technorati Tags: David Ferrucci, healthcare, IBM, Rod Smith, Watson

Dr. Eric Brown from IBM Research preps four members of congress - Jared Polis (D-Colo.), Bill Cassidy (R-La.), Jim Himes (D-Conn.), Rush Holt (D-N.J.) - for an exhibition game against IBM's Watson on Monday, Feb. 28, 2011. The match fostered a conversation among government leaders about the importance of IT to U.S. global competitiveness and encouraged greater focus on math and science education. Final score: Watson $40,300, Congressional Members $30,000. (Photo credit: Tom Briglia/Feature Photo Service for IBM)

Coming off the heels of its recent Jeopardy! win, IBM’s Watson computing system faced off against Congress last night for an exhibition match in Washington with five U.S.Congress members.

The bipartisan group put politics aside to test their trivia knowledge and foster conversations about the importance of IT to U.S. global competitiveness and encourage greater focus on math and science education.

The first of three practice rounds pitted Watson against Congressman Bill Cassidy, a Republican representing Louisiana’s 6th congressional district and Representative Rush Holt, a Democrat representing New Jersey’s 12th congressional district.  A doctor from Baton Rouge, Cassidy posted $1,000 to the board, but it was Representative Holt, a five-time Jeopardy! winner who gave Watson stiff competition.  Holt earned $8,600 against Watson’s $6,200 before the game progressed to Double Jeopardy.

Nan Hayworth (R-N.Y.), IBM's Watson, Jim Himes (D-Conn.), during an exhibition match in Washington

Next up, Representative Jared Polis from Colorado’s 2nd congressional district took on Watson and fellow house Democrat Jim Himes from Connecticut’s 4th district for the second practice round.  A Harvard graduate and Rhodes scholar, Himes relied heavily on his book smarts to post $7,600 and Polis stayed competitive with $6,800.  But Watson, a surprising fashionista, went on a tear during the “Always in Fashion” category racking up $22,500.

Bill Cassidy (R-La.), IBM's Watson, and Rush Holt (D-N.J.) during an exhibition match in Washington

With Watson’s victory close at hand, Representative Himes took to the podium again, undeterred in the third and final round.  Joined by Republican Representative Nan Hayworth, from New York’s 19th district — the birthplace of Watson where much IBM’s research and development was conducted — the two Northeastern members made a valiant effort before Watson closed out the final round of the exhibition match.

Final score:  Watson $40,300, Congressional Members $30,000.
But the informal game wasn’t all fun and games.  “This practice match is more than a trivia contest,” said Christopher Padilla, Vice President, IBM Governmental Programs.  “The technology behind Watson represents a major advancement in computing.  In the data intensive environment of government, this type of technology can help organizations make better decisions and improve how government helps its citizens.”

Technorati Tags: government, Watson

IBM's Watson computing system

When we think of the systems that make up a smarter planet, what typically comes to mind are industries like manufacturing, transportation, energy, or banking.  But there is another ‘industry’ that needs to become smarter.  We might call it the humanitarian industry.  That is, the system that creates a safety net to support society and is made up of philanthropies, social services, education organizations, NGOs and government agencies.

In many ways, this is the most human of all systems.  So it is ironic to consider how Watson, a computing system, could help us solve civic, social and cultural challenges and make smarter humanitarian decisions. But Watson’s deep QA technology presents new possibilities to do just that.  Through private sector collaboration with nonprofits, Watson can become the next innovation to be used as a force for societal good.

IBM has a strong track record in leveraging technology to help solve society’s challenges.  We’ve used grid technology to build World Community Grid, DNA analysis to develop the Genographic Project, and speech recognition technology to create Reading Companion.  What we’ve learned through these and other corporate citizenship efforts is that innovation can lead to outcomes with greater societal impact than simple corporate philanthropy alone (i.e. check writing).  We’ve also learned that getting the most from innovations like Watson requires collaboration and partnerships.

That’s why IBM recently invited about 100 leaders from the humanitarian sector to learn about the technology behind Watson and to discuss how we might work together to apply that technology toward achieving societal good.  Ideas ranged from how Watson could help us cut error rates of emergency systems (like 911 and 311) to how Watson might be used as an aid to transform how teachers teach and children learn.  You can learn more about the event by reading this article in Fast Company.

At the end of the event, we encouraged attendees to share their best ideas about how Watson’s technology might be applied to humanitarian endeavors.  I invite you to do so as well.  Please visit the citizen IBM blog to learn more and share your ideas.

Technorati Tags: citizenIBM, citizenship, humanitarian, nonprofit, philanthropy, Watson

From the IBM Watson Research Team: During Watson’s participation in Jeopardy! last week, we received a large number of questions (especially on reddit!) about Watson, how it was developed and how IBM plans to use it in the future. Below are answers to some of the most popular questions (you can also read the responses on the reddit blog). As background, here’s who’s on the team.

1. Could you give an example of a question (or question style) that Watson always struggled with? (Chumpesque)

Any questions that require the resolution of very opaque references especially to everyday knowledge that no one might have written about in an explicit way. For example, “If you’re standing, it’s the direction you should look to check out the wainscoting.”

Or questions that require a resolving and linking opaque and remote reference, for example “A relative of this inventor described him as a boy staring at the tea kettle for an hour watching it boil.”

The answer is James Watt, but he might have many relatives and there may be very many ways in which one of them described him as studying tea boil. So first, find every possible inventor (and there may be 10,000′s of inventors), then find each relative, then what they said about the inventor (which should express that he stared at boiling tea). Watson attempts to do exactly this kind of thing but there are many possible places to fail to build confident evidence in just a few seconds.

2. What was the biggest technological hurdle you had to overcome in the development of Watson? (this_is_not_the_cia)

Accelerating the innovation process – making it easy to combine, weigh evaluate and evolve many different independently developed algorithms that analyze language form different perspectives.

Watson is a leap in computers being able to understand natural language, which will help humans be able to find the answers they need from the vast amounts of information they deal with everyday. Think of Watson as a technology that will enable people to have the exact information they need at their fingertips.

3. Can you walk us through the logic Watson would go through to answer a question such as, “The antagonist of Stevenson’s Treasure Island.” (Who is Long John Silver?) (elmuchoprez)

Step One: Parses sentence to get some logical structure describing the answer

X is the answer.

antagonist(X).

antagonist_of(X, Stevenson’s Treasure Island).

modifies_possesive(Stevenson, Treasure Island).

modifies(Treasure, Island)

Step Two: Generates Semantic Assumptions

island(Treasure Island)

location(Treasure Island)

resort(Treasure Island)

book(Treasure Island)

movie(Treasure Island)

person(Stevenson)

organization(Stevenson)

company(Stevenson)

author(Stevenson)

director(Stevenson)

person(antagonist)

person(X)

Step Three: Builds different semantic queries based on phrases, keywords and semantic assumptions.

Step Four: Generates 100s of answers based on passage, documents and facts returned

from 3. Hopefully Long-John Silver is one of them.

Step Five: For each answer formulates new searches to find evidence in support or

refutation of answer — score the evidence.

Positive Examples:

Long-John Silver the main character in Treasure Island…..

The antagonist in Treasure Island is Long-John Silver

Treasure Island, by Stevenson was a great book.

One of the great antagonists of all time was Long-John Silver

Richard Lewis Stevenson’s book, Treasure Island features many great

characters, the greatest of which was Long-John Silver.

Step Six:  Generate, get evidence and score new assumptions

Positive Examples: (negative examples would support other characters,

people, books, etc associated with any Stevenson, Treasure or Island)

Stevenson = Richard Lewis Stevenson

“by Stevenson” –> Stevenson’s

main character –> antagonist

Step Seven: Combine all the evidence and their scores

Based on analysis of evidence for all possible answer compute a final

confidence and link back to the evidence.

Watson’s correctness will depend on evidence collection, analysis and

scoring algorithms and the machine learning used to weight and combine the

scores.

4. What is Watson’s strategy for seeking out Daily Doubles, and how did it compute how much to wager on the Daily Doubles and the final clue? (AstroCreep5000)

Watson’s strategy for seeking out Daily Doubles is the same as humans — Watson hunts around the part of the grid where they typically occur.

In order to compute how much to wager, Watson uses input like its general confidence, the current state of the game (how much ahead or behind), its confidence in the category and prior clues, what is at risk and known human betting behaviors. We ran Watson through many, many simulations to learn the optimal bet for increasing chances of winning.

5. It seems like Watson had an unfair advantage with the buzzer. How did Jeopardy! and IBM try to level the playing field? (Raldi)

Jeopardy! and IBM tried to ensure that both humans and machines had equivalent interfaces to the game. For example, they both had to press down on the same physical buzzer.  IBM had to develop a mechanical device that grips and physically pushes the button. Any given player however has different strengths and weakness relative to his/her/its competitors. Ken had a fast hand relative to his competitors and dominated many games because he had the right combination of language understanding, knowledge, confidence, strategy and speed. Everyone knows you need ALL these elements to be a Jeopardy! champion.

Both machine and human got the same clues at the same time — they read differently, they think differently, they play differently, they buzz differently but no player had an unfair advantage over the other in terms of how they interfaced with the game. If anything the human players could hear the clue being read and could anticipate when the buzzer would enable. This allowed them the ability to buzz in almost instantly and considerably faster than Watson’s fastest buzz. By timing the buzz just right like this, humans could beat Watson’s fastest reaction. At the same time, one of Watson’s strengths was its consistently fast buzz — only effective of course if it could understand the question in time, compute the answer and confidence and decide to buzz in before it was too late.

The clues are in English — Brad and Ken’s native language; not Watson’s. Watson analyzes the clue in natural language to understand what the clue is asking for. Once it has done that, it must sift through the equivalent of one million books to calculate an accurate response in 2-3 seconds and determine if it’s confident enough to buzz in, because in Jeopardy! you lose money if you buzz in and respond incorrectly. This is a huge challenge, especially because humans tend to know what they know and know what they don’t know. Watson has to do thousands of calculations before it knows what it knows and what it doesn’t.  The calculating of confidence based on evidence is a new technological capability that is going to be very significant in helping people in business and their personal lives, as it means a computer will be able to not only provide humans with suggested answers, but also provide an explanation of where the answers came from and why they seem correct.

6. What operating system does Watson use? What language is he written in? (RatherDashing)

Watson is powered by 10 racks of IBM Power 750 servers running Linux, and uses 15 terabytes of RAM, 2,880 processor cores and is capable of operating at 80 teraflops. Watson was written in mostly Java but also significant chunks of code are written C++ and Prolog, all components are deployed and integrated using UIMA.

Watson contains state-of-the-art parallel processing capabilities that allow it to run multiple hypotheses – around one million calculations – at the same time. Watson is running on 2,880 processor cores simultaneously, while your laptop likely contains four cores, of which perhaps two are used concurrently. Processing natural language is scientifically very difficult because there are many different ways the same information can be expressed. That means that Watson has to look at the data from scores of perspectives and combine and contrast the results. The parallel processing power provided by IBM Power 750 systems allows Watson to do thousands of analytical tasks simultaneously to come up with the best answer in under three seconds.

7. Are you pleased with Watson’s performance on Jeopardy!? Is it what you were expecting? (eustis)

We are pleased with Watson’s performance on Jeopardy!  While at times, Watson did provide the wrong response to the clues, such as its Toronto response, it is still a giant leap in a computer’s understanding of natural human language; in its ability to understand what the Jeopardy! clue was asking for and respond with the correct response the majority of the time.

8. Will Watson ever be available public [sic] on the Internet? (i4ybrid)

We envision Watson-like cloud services being offered by companies to consumers, and we are working to create a cloud version of Watson’s natural language processing. However, IBM is focused on creating technologies that help businesses make sense of data in order to enable companies to provide the best service to the consumer.

So, we are first focused on providing this technology to companies so that those companies can then provide improved services to consumers. The first industry we will provide the Watson technology to is the healthcare industry, to help physicians improve patient care.

Consider these numbers:

• Primary care physicians spend an average of only 10.7 – 18.7 minutes face-to-face with each patient per visit.
• Approximately 81% average 5 hours or less per month – or just over an hour a week — reading medical journals.
• An estimated 15% of diagnoses are inaccurate or incomplete.

In today’s healthcare environment, where physicians are often working with limited information and little time, the results can be fragmented care and errors that raise costs and threaten quality. What doctors need is an assistant who can quickly read and understand massive amounts of information and then provide useful suggestions.

In terms of other applications we’re exploring, here are a few examples of how Watson might some day be used:

• Watson technology offered through energy companies could teach us about our own energy consumption.  People querying Watson on how they might improve their energy management would draw on extensive knowledge of detailed smart meter data, weather and historical information.
• Watson technology offered through insurance companies would allow us to get the best recommendations from insurance agents and help us understand our policies more easily. For our questions about insurance coverage, the question answering system would access the text for that person’s actual policy, the other policies that they might have purchased, and any exclusions, endorsements, and riders.
• Watson technology offered through travel agents would more easily allow us to plan our vacations based on our interests, budget, desired temperature, and more. Instead of having to do lots of searching, Watson-like technology could help us quickly get the answers we need among all of the information that is out there on the Internet about hotels, destinations, events, typical weather, etc, to plan our travel faster.

9. How raw is your source data? I am sure that you distilled down whatever source materials you were using into something quick to query, but I noticed that on some of the possible answers Watson had, it looked like you weren’t sanitizing your sources too much; for example, some words were in all caps, or phrases included extraneous and unrelated bits. Did such inconsistencies not cause you any problems? Couldn’t Watson trip up an answer as a result? (knorby)

Some of the source data was very messy and we did several things to clean it up.  It was relatively rare, less than 1% of the time that this issue overtly surfaced in a confident answer. Evidentiary passages might have been weighed differently if they were cleaner, however. We did not measure how much of problem messy data effected evidence assessment.

10. I’m interested in how Watson is able to (sometimes) use object-specific questions like “Who is –” or “Where is –”. In the training/testing materials I saw, it seemed to be limited to “What is–” regardless of what is being talked about (“What is Shakespeare?”), which made me think that words were only words and Watson had no way of telling if a word was a person, place, or thing. Then in the Jeopardy challenge, there was plenty of “Who is–.” Was there a last-minute change to enable this, or was it there all along and I just never happened to catch it? I think that would help me understand the way that Watson stores and relates data. (wierdaaron)

Watson does distinguish between people, things, dates, events, etc. certainly for answering questions. It does not do it perfectly of course, there are many ambiguous cases where it struggles to resolve.  When formulating a response, however, since “What is….” was acceptable regardless, early on in the project, we did not make the effort to classify the answer for the response. Later in the project, we brought more of the algorithms used in determining the answer to help formulate the more accurate response phrase. So yes, there was a change in that we applied those algorithms, or the results there-of, to formulate the “who”/”what” response.

11. Now that both Deep Blue and Watson have proven to be successful, what is IBM’s next “great challenge”? (xeones)

We don’t assign grand challenges, grand challenges arrive based on our scientists’ insights and inspiration. One of the great things about working for IBM Research is that we have so much talent that we have ambitious projects going on in a wide variety of areas today. For example:

• We are working to make computing systems 1,000 times more powerful than they are today from the petascale to the exascale.
• We are working to make nanoelectronic devices 1,000 times smaller than they are today, moving us from an era of nanodevices to nanosystems. One of those systems we are working on is a DNA transistor, which could decode a human genome for under $1000, to help enable personalized medicine to become reality.
• We are working on technologies that move from an era of wireless connectivity — which we all enjoy today — to the Internet of Things and people, where all sorts of unexpected things can be connected to the Internet.

12. Can we have Watson itself / himself do an AMA? If you give him traditional questions, ie not phrased in the form they are on jeopardy, how well will he perform- how tailored is he to those questions, and how easy would it be to change that? Would it be unfeasible to hook him up to a website and let people run queries?

At this point, all Watson can do is play Jeopardy and provide responses in the Jeopardy format. However, we are collaborating with Nuance, Columbia University Medical Center and the University of Maryland School of Medicine to apply Watson technology to healthcare. You can read more about that here: http://www-03.ibm.com/press/us/en/pressrelease/33726.wss

13. After seeing the description of how Watson works, I found myself wondering whether what it does is really natural language processing, or something more akin to word association. That is to say, does Watson really need to understand syntax and meaning to just search its database for words and phrases associated with the words and phrases in the clue? How did Waston’s approach differ from simple phrase association (with some advanced knowledge of how Jeopardy clues work, such as using the word “this” to mean “blank”), and what would the benefit/drawback have been to taking that approach?

Watson performs deep parsing on questions and on background content to extract the syntactic structure of sentences (e.g., grammatical and logical structure) and then assign semantics (e.g., people, places, time, organization, actions, relationship etc).  Watson does this analysis on the Jeopardy! clue, but also on hundreds of millions of sentences from which it abstracts propositional knowledge about how different things relate to one another. This is necessary to generate plausible answers or to relate an evidentiary passage to a question even if they are expressed with different words or structures. Consider more complex clues like: “A relative of this inventor described him as a boy staring at the tea kettle for an hour watching it boil.”

Sometimes, of course, Jeopardy questions are best answered based on the weight of a simple word associations.  For example, “Got ___ !” – well if “Milk” occurs mostly frequently in association with this phrase in everything Watson processed, then Watson should answer “Milk”.  It’s a very quick and direct association based on the frequency of exposure to that context.

Other questions require a much deeper analysis. Watson has to try many different techniques, some deeper than others, for almost all questions and all at the same time to learn which produces the most compelling evidence. That is how it gets its confidence scores for its best answer. So even the ones that might have been answered based on word-association evidence, Watson also tried to answer other ways requiring much deeper analysis. If word association evidence produced strong evidence (high confidence scores) then that is what Watson goes with. We imagine this is to the way a person might quickly peruse many different paths toward an answer simultaneously but then will provide the answer they are most confident in being correct.

14. In the time it takes a human to even know they are hearing something (about .2 seconds) Watson has already read the question and done several million computations. It’s got a huge head start. Do you agree or disagree with that assessment?

The clues are in English — Brad and Ken’s native language; not Watson’s. Watson must calculate its response in 2-3 seconds and determine if it’s confident enough to buzz in, because as you know, you lose money if you buzz in and respond incorrectly. This is a huge challenge, especially because humans tend to know what they know and know what they don’t know. Watson has to do thousands of calculations before it knows what it knows and what it doesn’t.  The calculating of confidence based on evidence is a new technological capability that is going to be very significant in helping people in business and their personal lives, as it means a computer will be able to not only provide humans with suggested answers, but also provide an explanation of where the answers came from and why they seem correct. This will further human ability to make decisions.

Technorati Tags: Watson