About Keith

Keith Brawner currently works in the simulation industry for the DoD, before, during, and after getting a Masters in Intelligent Systems. Sadly, he is not yet a Doctor.

Tuesday, December 28, 2010

Self discovery enables robot social cognition: Are you my teacher?

"Self discovery enables robot social cognition: Are you my teacher?" was written by Kaipa, Bongard, and Meltzoff with the Universities of Vermont and Washington.  It can be found for free online here.


Life
First, sorry for the lack of blog posts.  I've not stopped reading papers, but the motivation to post things is low.  Additionally, my time is coming at a premium during the holiday season.  However, I didn't say that I was going to stop posting for a while, so I'm kind of a jerk.

Christmas was excellent, and I got to spend some time with my family and loving wife.  She is the best Christmas gift ever, besides maybe a briefcase.

Problem/Background
As they say, imitation is the finest form of flattery.  However, the question for a robot is who to imitate.  Work done in the field of robotics on how to learn from watching has been performed by Breazeal/Scassellati (2002) and Dautenhan/Nehaniv(2002), but the question of who to imitate is currently unsolved.  In 1997, Schaal showed 30 seconds of pole balancing video o a robot, and this was shown to increase learning rates.  This was one of the earlier experiments in this field, with more work done each year.  The paper does a great job of summarizing.



Experiment
The goal in this experiment is to have a mobile robot:
  1. Figure out what joints it has, and how it is able to move
  2. Figure out what other robot is similar enough to make a good teacher
  3. Figure out what individual actions imitate the teacher
  4. Actually perform said imitation
 
Self Discovery
There has been prior work done in self-discovery.  The robot is told of the part that make it up: how much each joint can move, and the shape/size/density of the materials that make it up.  The robot discofers possible position combinations through the technique used in Resilient Machines Through Continuous Self-Modeling (Bongard/Zykov/Lipson(2006)).
 
Self Modeling
A hill climbing algorithm is used to search the space of self models after all of the min/max parameter values have been found.  The self model is developed using a single genome.  This is then mutated (random -directioned Gaussian shift) and evaluated for error.  Lower error models are kept while higher error models are thrown away.

Teacher Modeling
Left and right cameras were used to capture data of a teacher robot using the process that Kaipa/Bongard/Meltzoff used in Combined Structure and Motion Extraction from visual data using Evolutionary Active Learning (2009).
 
Teacher Imitation
A 3x2 neural network is used to control the output of the student motors.  The output of the teacher is mapped to a hypothetical output of the student, and the network weights are trained.  The question of which of the teacher nodes map to the student nodes is solved during error reduction function.  It is worthwhile to note that because of the error reduction model, the student never performs the exact maneuver that the teacher does.

Future Work (and why you care)
Very simple models were used here, but many of them were used in series in order to perform the task.  Whenever the choice of optimization was presented, hill climbing was used.  Simple joints were used, and only two of them at that.  Simple cameras, teacher, etc. were used.  However the conclusion that a robot which doesn't even know it's own body may be able to learn from another robot or human teacher is a powerful conclusion.

Saturday, November 6, 2010

Neural Network Applications in Ship Research with Emphasis on the Identification of Roll Damping Coefficient of a Ship

Neural Network Applications in Ship Research with Emphasis on the Identification of Roll Damping Coefficient of a Ship was written out in Glasgow, Scotland's Department of Naval Architecture and Marine Engineering as a postdoctoral research paper by Mukhtiar Ali Unar.  It can be found for free online here.


Life
Still having a good time.  Still trying to get a job with the Army.  Also, my team sucks at trivia (we came in dead last, but regularly take 2nd or 3rd in the Thursday night games.
Problem
This paper is mainly a survey paper, but in brief, the main problem was to predict hydrodynamic coefficients for ship vessels.  The hydrodynamic coefficients are used to determine exactly how to execute the appropriate maneuver.  Due to a large number of factors, these cannot be directly measured and have thus far yielded to the 'best guess' of the people who are driving the boat.  In theory, if we were able to model the parameters better, we would be able to build safer and more reliable ships. 

Background
Artificial Neural Networks (ANNs or NNs) are a computational construction of a function that has many inputs and outputs.  It is a way to represent a problem as a 'black box', where the internals are unknown, but the inputs and outputs known.  The output of a neuron is the sum of the inputs to it, multiplied by their corresponding weights.  These weights can be 'trained' (adjusted in small increments) until they correctly model a given function.  It is represented to the right.

With that said, it is proven that any linearly separable function can be modeled by a Multi-Layer Perceptron (MLP) Neural Network (below).  It is represented below.




 








NNs have been in use (in research) for naval application since 1998 or so, with over 100 papers published using NNs in 2005.  They are a reasonably standard way of representing an unknown problem.


Naval Uses
  • Ship design (length, breadth, speed, draft, depth, displacement) - Clausen (2001)
  • Stability parameters (shipping vessels) - Alkan (2004)
  • Hull weight estimation - Wu (1999)
  • Estimation of wave induced ship hull bending - Xu/Haddarra (2001)
  • Automatic hull form generation - Islam (2001)
  • Control (steering, rudder roll, fin stabilization, collision avoidance, path following) - Various
  • Classification (filtering radar picture, marine acoustic signals, wake detection, ship trail clouds) - Various
  • Prediction of waves, tidal levels, storm surges, coastal water levels, and ocean currents - Various (1994-2007)

Experiment
So the goal here was to estimate ship parameters.  For this, a ship controller was constructed and fed data from a centralized database.  It was the job of the NN to stabilize the ship.  The NN error was charted, but not in a meaningful way (sorry Dr. Unar), as it seems like this paper was written for funding, rather than results.

 
Why do you care? 
Let's be honest: you probably don't.  The important takeaway here is that NNs are being used well outside of the traditional domain of artificial intelligence.  Although developed as an application of Machine Learning methods, they have now made their way into fairly mainstream applications.  Just remember that when you are looking at what a predicted storm surge for an area is, or how your semi-automated ship is turning, you are seeing the fallout of AI.

Sunday, October 31, 2010

Affect and Trust

Affect and Trust was written out of the College of Information Science and Technology in Drexel University, PA, by Lewis Hassel.  It can be found for free online here.


Life
It has been an exciting week, with a work party, Halloween party, Halloween outing (and old movie night) at the Enzian, and tonight's actual Halloween.  I think that with a party, movie, trick-or-treating, and Halloween Horror Nights, we may have bled this year dry.  Career-wise, RDECOM (now a subcomponent of ARL) has opened a position for me, and I have applied.  I can't wait to get my hands dirty with a research organization.  Of course, I am still reading Building Intelligent Interactive Tutors.

Issue
Hassel argues that there is a very distinct difference between trust and belief.  In his model, trust is based on action, while belief is based cognitively.  The example that shows this the most clearly is in the person falling backward.  He may believe that the other person is going to catch him, which is based logically on evidence that the other person is strong, not likely to want to inflict him harm, in close proximity, etc.  However, Hassel argues that he does not trust the person unless he actually falls.  While we may believe that we trust someone, we do not until the actual action is taken.

When do we trust someone?
Bos et al showed in 2002 that we tend to trust people more when meeting them face to face
Zheng has showed that we can trust people we've never met just as well, but that it takes long, and that just seeing a picture of the person is immensely helpful.  In fact, it is more helpful than seeing a datasheet of the person.

How do we model it?
note - PEU is Perceived Ease of Use
Each of the terms here is defined in the paper, as is each numbered path


Conclusion and Why do you care? 
It is important to know why you trust someone and why they trust you.  As we develop more advanced models of the phenomenon, we understand it better.  As we understand it better, we learn how we can trust others more, and how we can build our own trust with others.

Sunday, October 24, 2010

Rant - AI in Education - Building Intelligent Tutors

I am currently reading Building Intelligent Interactive Tutors: Student-centered strategies for revolutionizing e-learning.  You can buy the book on Amazon here.  For, you can download it, say, for an ebook reader, here.

Life
Super-awesome weekend.  We went to Halloween Horror Nights at Universal Studios to see movie-quality, real life, monsters that jump out at you in a number of haunted houses (8? 10?), combines with the great rides of a world-class theme park.  In addition to that splurge of entertainment, we went to La Nouba, and were thoroughly entertained by live performers for over an hour.  In a few hours, we will be attending a block party.  So, please forgive the late update.


Breaking points
     I've been reading Building Intelligent Interactive Tutors by Beverly Park Woolf, and one of the things that she speaks of often is the idea that each industry reaches a critical threshold on occasion.  For instance, the field of computer science benefits from object oriented programming/design.  The field of physics has made leaps and bounds based on the models that they can now create via computer simulation.  She argues that the field of education is now overdue for such a breakthrough for a few reasons.  In fact, just this month this subject was a featured article on the technology site Slashdot.  You can read more here.


Why now?

In the past field of education, learning has been studied, and segmented into a few categories:
  • one-on-one instruction versus group (one-on-one is significantly more effective)
  • inquiry learning versus lecture learning (inquiry is more effective)
  • Testing versus teaching (tests can make ability gaugeable, but the time is better spent teaching if you already know the ability level)
  • motivational learning versus subject learning (students learn better when motivated)
  • Mastery learning (building a subject from the ground up, and asking 'why?') outperforms other forms of learning.
Logically speaking, you want a one-on-one teacher that teaches via asking questions (or better yet, getting students to ask the right questions), without any tests, in a subject that the student is interested in.  I can see you rolling your eyes at this.  Despite knowing that these teaching methods are the most effective, they are also the most difficult to implement.  Having one first grade teacher per student is ludicrous, and attempting to get them to sit still long enough to actually ask questions about subject matter isn't exactly realistic.
Or is it?
There is an obvious exception to this, however, and my reader likely sees it coming.  Intelligent Tutoring Systems offer the real promise of optimal learning.  With each of these subject-area improvements, you can make leaps and bounds with performance.
  • ITS's can tutor one-on-one, and are best this way
  • ITS's can teach via inquiry learning, either by providing a large number of questions, or by grammar-parsing text-written (or spoken) response
  • An ITS has no real need to test.  When working a domain like mathematics, it can assign homework problems that are graded on-spot.
  • ITS's can gauge student involvement as well as or better than a live tutor, using sensors
  • ITS's can use Mastery Learning if constructed in the correct manner by an expert (say, a grade school teacher).

Why do you care?
There is a strong case to be made that the students of the future will be taught via a computer interface that is customized to their needs.  It will keep track of their learning on various subjects, get their interest and keep it, and get them to ask questions about the subject matter.  It is likely that it will be able to be distributed via Internet, and that a large portion of mankind will be bettered by it.  People in first world countries will be getting the same education that a significant portion of the planet is getting.
There are still some important problems to solve (for instance, all of the above), but it is likely to be only a matter of time before they can be taken care of.

Friday, October 15, 2010

Predicting Searcher Frustration

Predicting Searcher Frustration was written out of the University of Massachesetts by Feild, Allan, in conjunction with Jones from Yahoo!.  It can be found for free online here.

Life
The plan for this weekend is to go camping and enjoy some time out-of-doors.  There are plans for s'mores, canoeing, and hiking, in addition to whatever else strikes our fancy.

Problem
Search engines, like all businesses, are striving to be better in order to claim more market shares, ad revenue, and viewers.  As part of this effort, one of the things that they (or at least Yahoo!) are looking into is predicting when users are getting tired/frustrated in looking for data.  If they detect that a specific user is frustrated, then presumably they could make the interface better, give different results, give a different category of results, or simply mark it as an area for future improvement.


Experiment
What we are going to do here is make a bunch of users go on a scavenger hunt for information, and report how they feel about it.  This will be measured in a couple of different ways:
  • Query Logs - including page focuses, clicks, navigation, mouse movements, etc. (47 features in total)
  • Sensor Data - including a mental state camera, pressure sensitive mouse, and pressure sensitive chair
    • mental state camera has 6 states - agree disagree, unsure, interested, thinking, confident
    • mouse has 6 pressure sensors - 2 on top, 2 on each side
    • chair has six sensors - 3 on back, 3 on seat

Results
As you can see from the right, there are a few important conclusions:
Why do you care? 
1 - Search engines are getting better, and user modeling is likely to play a role in this in the future
2 - Direct sensor data is not required in order to predict how you are feeling (your webpage views alone are more accurate)

Friday, October 8, 2010

“Yes!”: Using Tutor and Sensor Data to Predict Moments of Delight during Instructional Activities

“Yes!”: Using Tutor and Sensor Data to Predict Moments of Delight during Instructional Activities was written out of Arizona State by Muldner, Burleson, and VanLehn.  It can be found for free online here.

Life
Nothing particularly fancy is going on.  The combined Regular Day Off and Columbus Day have granted a glorious 4-day weekend.  This combined with today's high of 85 will make for a relaxing weekend of light reading, sushi, and entertainment.

Career
I/ITSEC, the international conference on modeling and simulation will be in Orlando next month.  It will be a good time, offering everything from paper presentations on how to train nurses, to the latest advancements in computer graphics, to 3 dimensional printing, to firing grenade launchers (mockups with recoil) at insurgents (simulated).

Problem

The AI in Eduation field, in general, has been focusing on trying to keep students 'in the zone', or to keep them from getting frustrated through the use of hints, easy questions, scaffolding, or a number of other methods.  However, this paper postulates that the most important moments in education are the "yes!" moments or great success.  However, we don't know how to detect these moments and may interfere with their occurrence.


"Yes!"
Most everyone has had a "yes!" moment in their education.  If you think, this is the moment where you had a sudden realization of a concept, or when you had just answered a particularly difficult problem.  This moment probably involved significant work with sudden reward.  As my high school calculus teacher would say "These are the moments when the student finally understands, and the moments I live for."

Experiment
"Yes!" data was gathered using the Example Analogy (EA) Coach with Newtonian physics.  Interactions with the interface were recorded, and students were asked to think aloud.  Additionally, a posture chair, skin-conductance bracelet, eye tracker, and pressure mouse were used to gather data on the students current state.  The "yes!" moment was labeled by an expert, and the system trained to recognize the occurrence based upon that information.

Results
Posture Chair Data
Logistic regression was used to attempt to make sense out of the sensor data.  As has been found in other studies (such as Automatic prediction of frustation), the data from the posture chair was not particularly usable.  However, through the use of time-based models which included pupil response and imput from the other sensors, they were able to correctly predict 60% of the "yes!" events, while incorrectly predicting non-"yes!" events 13% of the time.  Obviously there is some work left to be done in the field, but these results are promising and show possibility.

Why do you care? 
1 - ITS systems can keep students optimally challenged if they are reporting a high frequency of "yes!" events.  This is just as important, if not more important, than predicting frustration.
2 - Detection of such events is possible, and should be further investigated.

Friday, October 1, 2010

Exploring the Possibility of Using Humanoid Robots as Instructional Tools for Teaching a Second Language in Primary School

Exploring the Possibility of Using Humanoid Robots as Instructional Tools for Teaching a Second Language in Primary School was written out in National Central University (Taiwan) by Chang, Lee, Chao, Wang, and Chen.  It can be found for free online here (page 18)).


Life
Mostly just chilling at home, and hanging out with friends (trivia, D&D, dinner), and going to work.  All is clear on this eastern front.  Oh yea, and we've been goofing around with the Sony 900BC eReader, which is proving itself useful for reading Dune, Ghost in the Shell, and, of course, more research papers!


Problem

This paper is unusual in the sense that it doesn't actually propose a problem.  The problem is well studied: people learn foreign languages poorly.  With that said, we will study the availability of specific tools to tackle it.  Today's tool: Robots.


Experiment
Our research teacher in high school was fond of saying that you are not to say "I hypothesize that RAID will kill roaches".  Instead, you are to say "I hypothesize that the addition of RAID to roaches will have a measurable effect with regard to activity, food intake, etc.".  For those of us watching carefully, all we said there was that we thing something will happen, and will measure some stuff to see if something happened.  Today we think that robots will have a net benefit effect on the classroom.

Why and How?
Robots have some of the benefits listed to the left, with regards to teaching.  The researchers here made 5 modes of operation in order to attempt engagement:
  • Storytelling - robots tells a story in a foreign language, complete with different voices for characters
  • Oral Reading - Robot reads a printed story aloud, and calls upon children to help it
  • Cheerleader - Robot encourages students to participate in games, and does dances when students get the answers right
  • Action Command - Robot plays Simon Says with the students
  • Question-and-answer - Robot talks to students, asks them to introduce themselves, introduces itself, etc.  Robot plays the role of a foreign persona (male or female)
Results
When there is a robot in the class (doing the above):
  • Students respond more loudly, speak more often, ask more questions, listen quieter, and watch the robot intensely (according to teacher survey)
  • Shier students interact more, while more outgoing students interact for longer periods of time
  • Teachers reported offloading of teaching as a robot can perform roles of either sex while the teacher is limited to one, mostly.
  • Teacher had additional time to work with poorer students while everyone was distracted with a robot. 
Issues
  • They report lack of training
  • Complicated technology 
  • Decaying motivation
  • Robot shows no emotion
Conclusion:
Robots are awesome.  When students are talking to a robot they are more engaged, and talk/practice their language skills more often.  This technology can be coupled with Affective Tutoring in order to further help the learner.

Why do you care? 
Coming in at a whopping price of $250 (or roughly 5 textbooks), we may see the involvement of robotic helpers in the school system relatively soon.  These advances, coupled with the advancement of image/face recognition could result in genuine custom tutoring available for relatively cheap.

Friday, September 24, 2010

Investigating the Relationship between Presence and Learning in a Serious Game

Investigating the Relationship between Presence and Learning in a Serious Game was written out of the Institute for Creative Technologies in University of Southern California by Lane, Hays, Auerbach, and Core.  It can be found for free online here.


Life
I am currently back in Orlando Florida for a bit of a short/relaxing week.  I am coming down with a cold, and my step-sister Crissie is getting married on Sunday.


Problem
Apparently when people experience 3D events they are actually able to change their frame of reference, and experience the event as reality (see: Narrative  impact : social and cognitive foundations).  Furthermore, fantasy concepts are found to be more helpful for teaching children (page 240).  However, studies with Crystal Island (a 3D immersive world) haven't shown that the 3D world helps.  This paper sets out to find out, and used BiLAT (a system for teaching Arabic culture).

Experiment
One group of students will be tutored in a 2D interface, while another is tutored in 3D.  We see how well they did based on Pre/Post-Test results.  Additionally, the students will be asked various questions about how involved in the training they were.  I'll gloss over the results of the second part and just tell you that the 3D people were more involved;  no surprise here.

Results
Posted here as a summary:
  • In-game, both groups made the same number of errors
  • With coaching (ITS), roughly the same number of errors were made
  • During post-test, the 3D students made less errors (but not statistically significant)
  • 2D students performed twice as many actions (clicks/responses) as 3D students
  • when coaching was not available, 3D students took more time to select answers than 2D students (suggesting that they treated it as a more real social interaction)
Conclusion: More Study needed


Why do you care? 
First, remember to teach your children with fairy tales, as they are shown to teach lessons better.  Additionally, there is evidence to suggest that your 3D video games make more of an impression than your older ones.  However, just because you are learning in 3D doesn't mean that you are learning more.

Friday, September 17, 2010

Affective Gendered Learning Companions

Affective Gendered Learning Companions was written out of the UMASS Amherst Comp Sci, Psych, and Education departments by Arroyo, Woolf, Royer, and Tai.  It can be found for free online here.


Life
I am currently posting from New London, CT, the submarine capitol of the world. Work-wise, we have recently completely an installation of submarine training software for both enlisted personnel and officers.


Problem
The short version here is that the field of mathematics is having some trouble getting girls on board for the big win.  There have been a number of studies devoted to the subject, but the answer is blunt: girls think that they aren't any good at math (whether they are or not) starting in middle school.  You know what they say: "Whether you think you can, or you think you can't: you're right".  Shortly after this time, test scores plummet.  Later, girls become somewhat rare in the Science, Technology, Engineering, and Mathematics (STEM) fields.  Ideally, we can improve these test scores by using an Intelligent Tutoring System (ITS).

Affective Tutoring
One of the core concepts in ITS development is the idea that people can get into a 'zone' where they are learning.  Hopefully you can sucker/bribe them into it and trap them.  This is usually an extension of the groundbreaking discoveries that "people don't learn well when they are tired" and "Little Timmy is angry right now, and won't sit still".  If you had a real tutor sitting in front of this person, they would try to calm the student down.  If the student is frustrated with a certain type of problem, you can switch problems.  If the student is learning a concept well, you let them.  Computer-based tutors are going to have to tackle the same issues.

Presumably, affective tutoring is better than non-affective and tutoring someone is better than not*, and there are a couple of challenges to managing student emotional state:
1 - Computers don't understand emotions
2 - People don't understand emotions
3 - People that program computers don't understand emotions
4 - It's not clear what the most beneficial response to "I hate math!" is for a math tutor (even when the emotion is clear).

Pedagogical Agents
Remember the little picture from the problem paragraph?  Yea, it's a little cartoon character that studies with you.  That's a pedagogical agent.

Experiment
So, we don't know how people actually feel (emotional state), but we can ask them!  Then, when they say that they are anxious, we can make the cartoon anxious too.  This way the learners can feel like there is someone suffering along with them.  It's cheap, but it works.

So here is the big question that all of that background built up to: Do boys/girls prefer boys/girls for their learning buddy?  We'll test some with each and some without.


Results
This paper did not inform on whether students performed better with or without a cartoon character (see Empathetic Pedagogical Agents by Arroyo, cartoons help) but had a rather important finding (below).  Girls who study with a girl character think that they are worse at math than when they started (they aren't, they tested better).  However, girls who study with a boy character think that they are better.



Why do you care? 
First, given that the character is a cartoon, and that they say the same thing, it is a bit surprising to find that there is even a statistical preference.  Also, even if there is a preference, the idea that the preference would affect self opinion, regardless of actual performance, is surprising.  Finally, if you have a daughter, make sure her tutor or study-partner is a boy.

* - Some studies show that you just want answers.  Here.

Administrative BS and Format

Yes, eventually this site will have some actual content.  Since I intend on doing this on a fairly regular basis, I figure that I will try to follow a formula.  If the formula sucks, I'll change it.  I'll try to keep it a bit on the snarky, and a bit on the short.

SECTION 1 - What is going on in my life.

SECTION 2 - Brief summary of what the paper is setting out to prove (Neural networks are an awesome construct of computing, etc.)

SECTION 3 - Any important field aspects of the paper (description of new Neural Network model that they are using, etc.)

SECTION 4 - What the paper accomplishes (Neural Networks changed the field of image recognition).

SECITON 5 - Why it matters, why you care (if it matters, if you care...)

Topic of the Day - AI in Education

It is no secret - I have applied for a job working for the US Army, in RDECOM Orlando.  As part of 'spinning up', I am reading roughly 1 paper/day on the subject of AI in Education hopes of starting work with an Intelligent Tutoring System (ITS) project.

As such, I'm afraid a bunch of the research papers summarized after this post will be in the AI-ED domain, although I'll try to spice things up with the occasional Neural Networks paper.

More information on AI in Eduction can be found below:
Journal of AI in Education
AI in Education 2009 Conference
Tenth International Conference on Intelligent Tutoring Systems

My personal keep-track of AI-EDU papers, here.

Founding Post, Mission Statements, and answers to basic questions

"If a man were to commit himself to bettering himself in his profession 1 hour per day, he would be a leading expert in his field within 2 years" - Unknown

The above quote was shared to me during a training class that I took quite some time ago, but I believe that it still holds true today.  So, let's get down to business and do what business people do: Make Mission Statements.

Why am I creating this blog?
To document the research that I am reading and make it more accessible to the public, for their benefit, in a timely manner.

Also, apparently when I don't have any homework, I assign myself some.

How often do I intend to update it?
Weekly.

That's an awful lot of updating, what are you going to talk about?
I am going to review research papers of my interests (AI, AI in Ed, AI in Signal Processing/Control, etc.).

Who the hell is going to read that?
Anyone who wants a quick summary on the absolute-latest-state-of-the-art, without necessarily wanting to get an advanced degree.

What about other things you are interested in, like Caving, Personal Finance, RPGs, or Travel?
They will either get another blog, or I'll do what I've always done: not post information on the Internet about them.

Do I have a PhD?
No, and I have currently stopped at my MS.  If I continue down my current path, however, I will have one before the end of 2013, but I've not started.

If you don't have a PhD, why did you call it Dr. Brawner's Research Blog?
Because it sounded catchy, brawner.blogspot.com and brawnerblog.blogspot.com were taken, and as a tip-of-the-hat to Joss Whedon.