Nabin K. Malakar, Ph.D.

NASA JPL
I am a computational physicist working on societal applications of machine-learning techniques.

Research Links

My research interests span multi-disciplinary fields involving Societal applications of Machine Learning, Decision-theoretic approach to automated Experimental Design, Bayesian statistical data analysis and signal processing.

Linkedin


Interested about the picture? Autonomous experimental design allows us to answer the question of where to take the measurements. More about it is here...

Hobbies

I addition to the research, I also like to hike, bike, read and play with water color.

Thanks for the visit. Please feel free to visit my Weblogs.

Welcome to nabinkm.com. Please visit again.

Saturday, June 30, 2012

Cracking the World's hardest Sudoku

I heard about the claim that this one is the hardest sudoku puzzle made to date. Obviously I wanted to solve it. But since I am pressed with time this moment, I solved it using Google's Goggles. :)




 -

Thursday, June 28, 2012

Rock paper scissors, wait there is more...

You can win the game (or battle) with few methods: with intelligence/tactics, and/or with speed (assuming equally powerful players).
The field of Machine learning can use both! :)
So, fasten your seatbelt and think!
Recently I saw a video showing robot system with 100% winning rate "as one example of human-machine cooperation systems". The key was the speed. http://www.k2.t.u-tokyo.ac.jp/fusion/Janken/index-e.html

On the lighter note to the post, because of limited outcomes of the Rock-paper-scissors, there might be draws. So there is more to the game, if you are willing to add the options. As BBT suggests: "Rock-paper-scissors-lizard-spock".

Monday, June 25, 2012

Forget about the Class, focus on learning

Rame failed in class 3, and had to repeat again.

Harke failed in class 6 twice, he had to go back to the same class while his friends were already in class 8.

Furke did so good in class 5 that he was promoted to class 7.

These are few examples of how we try to manage the student's ability to excel in course structures in school. Since the course structures are reformed once in a blue moon, a generation can suffer if it has not been well constructed.

Forget about the class system. My proposal is to make it in more flexible way.
It can be done in two ways:
1. Letter grade system: A student interested in Economics might not need to excel in another subject that he fails most of the time.
By making it mandatory that people hiring in Economics need students who got at least B in that subject, the competitive edge will evolve so that the students who favors particular subject will go in their desired direction. If you are a competitive company you would not hire an accountant who got E on account courses. (See second point on foucsed learning.)

2. Forget about the class system. As discussed on the opening line, students loose their precious years only because they had bad subject attitudes. We all have different degrees of affinity towards different subject lines. Modify the system it so that a student can take Algebra from class7, arithmetic from class 8 and geometry from class 6. However, in order for him/her to appear in the SLC, s/he must have the certification from all the basics upto "class" 10.  If s/he fails to do so, he can still choose to take CTEVT courses while enrolling part-time to the subject for the certificate of school. That way, the student need not study back in the same class for next year, and yet learn new subjects while refining the weak subject.

We all get ready to talk about the politics, and the next big things to do in #Naya Nepal... while, there are so many short-term things to do (and can be done) in Nepal. The short-term things are the stuffs that need to be done before we can achieve the big thing. Remember the whole is by part theorem?
Education sector is the one of them that needs some change. Your inputs are welcome.
>

Sunday, June 24, 2012

Embryonic development of Fruitfly, cell by cell

Embryonic development of fruit flies has been tracked in realtime by using multidirectional imaging technique.

While watching it, ask yourself when does the life enter into the clusters of cell...


More...
Fruitfly development, cell by cell : Nature News & Comment:

"References
Tomer, R., Khairy, K., Amat, F. & Keller, P. Nature Methods http://dx.doi.org/10.1038/nmeth.2062 (2012).
Show context
Krzic, U., Gunthur, S., Saunders, T. E., Streichan, S. J. & Hufnagel, L. Nature Methods http://dx.doi.org/10.1038/nmeth.2064 (2012)."

Thursday, June 7, 2012

FBI records of Feynman released to public

Richard Feynman, born on May 11, 1918 @ Far Rockaway, Queens, New York, U.S. was an American physicist.
He is a very well known figure for his work in the path integral formulation of quantum mechanics, the theory of quantum electrodynamics, and the physics of the superfluidity of supercooled liquid helium, as well as in particle physics...  [read more here].

At the height of cold war, he was investigated by FBI. Now the FBI records have been released to the public.
Follow the detailed story at:
http://www.muckrock.com/news/archives/2012/jun/06/feynman-files-professors-invitation-past-iron-curt/
>

Wednesday, June 6, 2012

शुक्रलाई ताकेर सूर्यलाई हेर्ने :)

म चाहि यो कुरोको पर्दाफास गर्ने प्रयत्नमा छु ।


नपत्याए यो भिडियो हेर्नुस त! करीब सात घंटा लामो शुक्रगमन को मौक़ा छोपेर खिचिएको यो भिडियोमा  एउटा थोप्लो वारि बाट पारी जान्छ, र त्येसै समयभित्र सूर्यमा कति चलखेल हुन्छ !
बहाना न हो, कुन मौक़ा पाऊ र कनिका बुकाऊ भन्ने दाऊ खोजी रहेका बिश्वभरीका बैज्ञानिकहरुले सुक्र्गमनको मौक़ा छोपेर सूर्यको बारेमा समेत सिक्षा दिए ।  यसैलाई "शुक्रलाई ताकेर सूर्यलाई हेर्ने" भन्ने रणनीति भन्नुपर्ला !
एक-जुगमा एक-दिन भन्थे, अब यो मौक़ा हाम्रो जुनीमा  फेरी फेरी आउदैन ।  


>











=== On June 5-6 2012, SDO is collecting images of one of the rarest predictable solar events: the transit of Venus across the face of the sun. This event happens in pairs eight years apart that are separated from each other by 105 or 121 years. The last transit was in 2004 and the next will not happen until 2117. Credit: NASA/SDO, AIA To read more about the 2012 Venus Transit go to: sunearthday.nasa.gov/transitofvenus

Monday, June 4, 2012

Google drive: a missed opportunity

No, I am not talking about whether launching Google Drive was too late. Nor whether it was up par with available services. Many people have expressed their views.

What I was thinking was in a research perspective. Now and then there are research about tracing the social influence map of the internet. While releasing the products like Google Drive, Google/ or any company gets a rare opportunity to map the influence or spread of it.  For example a news splash by techcrunch might be spread much faster than say another site with similar theme.
What Google could have done was incentive-tize the process of referring the Google Drive. For example, Dropbox and Sugarsync (yes they have referral links) give the referral links for people who want to recommend it to their friends/mass. Studying the click through rate via referral link, can be done by enabling invite only through Emails ("Give G-drive to your friend").

This is not only true for Google, but also to any other company who is planning to release a new product, that could go vibrating in the web.
My assumption is that by studying the spread of influence, it might be useful to infer how the news spread in the human network. Or, at the least, who is the most relevant networker to get the point across.

>

Saturday, June 2, 2012

Common Student Mistakes in Lab Report

One of the TA duties of graduate students is to instruct in Lab and grade the lab reports. The report writing aspect of lab is very useful skill, which students can learn and also find it useful later in their career.

Here is a list of common mistakes the students make in the lab report (in no particular order):

  • Use the wrong date (often use the date they do write-up rather than the date of the lab)
  • Forget their lab partners last/name
  • Wrong format for name, lab, etc
  • Introduce the lab rather than simply state their objective
  • Forget to define the equation variable in the theory
  • Forget to relate theory to experiment
  • Forget to put units (or wrong units)
  • Not state general equations
  • Error not 1 sig fig or rounded down
  • Value corresponding to error has more decimal places than the decimal place of the error
  • Forget to put title on the graph or have x vs y instead of y vs x
  • max/min lines not encompass all of the points and their error
  • Not labeling the y-intercepts and/or the slopes (or labeling incorrectly)
  • Majority of the graph space left unused
  • Use data points instead of graph points to find the slope
  • Forget to compare equation of a line with theory equation
  • Not restate results to prove statements in discussion and conclusion
  • Use human error such as "I may have calculated wrong" or "human reaction time" in the sources of error.

Would you like to add that you have encountered?

Adopted from
http://www.unbc.ca/assets/physics/labs/phys_labs_common_mistakes.pdf

>

Friday, June 1, 2012

Write your Name with Galaxies: Galaxify

Reach for the stars, write your name with Galaxies!
Note the website above, created by astronomer Steven Bamford, which uses images of galaxies to recreate your message as an image.  The galaxy images are acquired through the Sloan Digital Sky Survey and Galaxy Zoo projects.

Well, nabinkm.com appears as:

Did you ask: Which galaxies are used in this image?
























For the Geeky love birds out there:

Find out which ones were used!
>