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.

Friday, January 24, 2014

Physics Olympiad '14 Selection by Nepal Physical Society

KATHMANDU: Nepal Physical Society is conducting an entrance examination for national level physics Olympiad-2014 on February 8. 

The Nepal Physics Olympiad (NePho) Committee has been conducting the exams for the higher secondary level and equivalent science students for the last six years. NPS will provide intensive tutorial and training to 20 best students, of which five will be selected for the 45th International Physics Olympiad scheduled to be held from July 13 to 21 in Astana, Kazakhstan. “The final team will undergo another set of tutorial and training,” said Indra Bahadur Karki, Secretary of Associate Professor of Physics. Interested students should not be enrolled in any college and they should be below 20 years as of June 30, 2014. The registration forms for the entrance exam must be submitted before February 3.
(Published in Himalayan Times
http://www.thehimalayantimes.com/fullNews.php?headline=Physics+Olympiad+%2714&NewsID=403877

 NePhO Model Questions  (2013)

Contact NPS for the details
http://nps.org.np/

-->

Thursday, January 23, 2014

Assessing Surface PM2.5 Estimates Using Data Fusion of Active and Passive Remote Sensing Methods

In this paper, we focus on estimations of fine particulate matter by combining MODIS satellite Aerosol Optical Depth (AOD) with Weather Research Forecast (WRF) PBL information using a neural network approach and assess its performance. As part of our analysis, we first explore the baseline effectiveness of AOD and PBL as relevant factors in estimating PM2.5 in passive radiometer and active LIDAR data at CCNY and demonstrate that the PBL height is the most critical additional parameter for accurate PM2.5. Furthermore, active measurements from both ground and satellite based lidar are used to show that summer WRF model PBL heights are most accurate. We then expand our analysis to a regional domain where daily estimations are obtained and compared with operational GEOS-CHEM PM2.5 product. Using our approach, we also create regional daily PM2.5 maps and compare against GEOS-CHEM outputs. Finally, we also consider additional improvements, where multiple satellite observations are used as regressors to predict PM2.5. These results illustrate the significant improvement we obtain within this framework in comparison to a “one size fits all continental scale approach”.
PM2.5 estimation for NY and surrounding states for a particular day.
Published in British Journal of Environment and Climate Change, ISSN: 2231–4784 ,Vol.: 3, Issue.: 4 (October-December)-Special Issue
See full article at: http://www.sciencedomain.org/abstract.php?iid=323&id=10&aid=2530

Saturday, January 18, 2014

Survey On The Estimation Of Mutual Information Methods as a Measure of Dependency Versus Correlation Analysis

Link:
http://arxiv.org/abs/1401.3358

In this survey, we present and compare different approaches to estimate Mutual Information (MI) from data to analyze general dependencies between variables of interest in a system. We demonstrate the performance difference of MI versus correlation analysis, which is only optimal in case of linear dependencies. First, we use a piece-wise constant Bayesian methodology using a general Dirichlet prior. In this estimation method, we use a two-stage approach where we approximate the probability distribution first and then calculate the marginal and joint entropies. Here, we demonstrate the performance of this Bayesian approach versus the others for computing the dependency between different variables. We also compare these with linear correlation analysis. Finally, we apply MI and correlation analysis to the identification of the bias in the determination of the aerosol optical depth (AOD) by the satellite based Moderate Resolution Imaging Spectroradiometer (MODIS) and the ground based AErosol RObotic NETwork (AERONET). Here, we observe that the AOD measurements by these two instruments might be different for the same location. The reason of this bias is explored by quantifying the dependencies between the bias and 15 other variables including cloud cover, surface reflectivity and others.

And related:
Towards Identification of Relevant Variables in the observed Aerosol Optical Depth Bias between MODIS and AERONET observations

http://arxiv.org/abs/1302.2969


Estimation and bias correction of aerosol abundance using data-driven machine learning and remote sensing
http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6382197&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6382197
-->

Tuesday, January 7, 2014

Interview with Dr. Mim Lal Nakarmi

We are presenting a brief interview with Dr. Mim Lal Nakarmi.  He was recently promoted as a tenured professor in Brooklyn College, NY. Presenting a brief, yet inspiring interview with him.


Prof. Mim Lal Nakarmi
0. Sir, Congratulations on getting the promotion and Tenure @ Brooklyn College. Could you please tell us a little bit about yourself.  
MLN:   I was born and grown up in Banepa. All my school level learning was in Banepa. At college level, I did I.Sc. at ASCOL, B. Sc. at Tri-Chandra, and M. Sc. at TU, Kirtipur. After M.Sc. I started teaching at Kathmandu University. While teaching there, I also did M. S. in Electronics from BITS (Birla Institute of Technology and Sciences), Rajasthan, India. After teaching at Kathmandu University for some years, I started Ph.D. program in physics at Kansas State University (K-State) from 2000 and finished in 2005. After that I worked as a post-doc for two years in the same research group. In 2007, I moved to New York to start my career as tenure-track Assistant Professor at Brooklyn College of the City University of New York (CUNY). My tenure and promotion will be effective from Fall 2014.
My research field is experimental semiconductor physics. I am involved mainly in the growth and characterization of wide band gap semiconductors for optoelectronic applications. I spent most of the time in the development of semiconductor materials such as Aluminum Nitride (AlN) and Aluminum Gallium Nitride (AlGaN) alloys for deep UV applications during Ph.D. and post-doc. In this course, we have to study structural, electrical, optical properties of the materials. AlN that has a direct band gap of about 6.1 eV at room temperature is the one I explored the most and utilized in the fabrication of 280 nm deep UV light emitting diodes (LEDs). Deep UV LEDs have application in next generation general purpose lighting, air/water purification, bio-chemical agent detection, medical/research applications etc. I am continuing the similar research at Brooklyn College. Recently I collaborated to work on zinc oxide (ZnO) aiming to achieve p-type. I have also studied optical and electrical properties of multiferroic materials in collaboration with Prof. Ram Rai at SUNY College at Buffalo. I am building my research lab facility for material synthesis. My focus will still be on deep UV materials, but I am also going to develop different structures such as one dimensional nanowires and two dimensional mono-layer in addition to thin films structures. My research activities can be viewed in my websites http://www.brooklyn.cuny.edu/web/academics/faculty/faculty_profile.jsp?faculty=664Or http://userhome.brooklyn.cuny.edu/mlnakarmi/

1. What was your aim in life as a teenager? How did you decide to study physics? When did you know you wanted to be a physicist? Did anyone, in particular, influence you? 
 MLN:  I used to say I would be a doctor in future. But I joined in a physical group in I. Sc. because biology was not my favorite subject. So I was in engineering track sort of in I. Sc. but at the same time I found physics a very interesting subject.  So, my interest to be a physics student ought to be seeded during the I. Sc. period. Our family has already an engineer and I was not attracted that much in that field. Without seeking aggressively for engineering admission, I did B.Sc. and M.Sc. with physics and math. Although those years were like roaming without precise destination, after I started teaching at KU as a Lecturer, I came to realize that there is no real future without Ph.D. in that career. That time, few people have already started Ph.D. programs abroad. I also started seeking for Ph.D. admission. Email/internet was just made available in Nepal. That helped me a lot to get information about admission for PhD. I am very thankful to my friends Sunil Shrestha and Jagat Shakya who were already in USA at that time for my admission in PhD.

2. What strategies did you use to be successful in college, as a student? 
MLN:   I was not a very good student at colleges in Nepal. I like to study to understand the subject. Since the exams in Nepal are not the test of understanding, my exam scores were not good. In other words exams in Nepal are not test of knowledge rather memory test. So my habit and strategy did not work. But I did much better when I studied at BITS and K-State because they look for conceptual understanding and test in the same way in the exams. For the classroom performance, I also try to understand teacher’s psychology. If you have experience of teaching and writing exams it’s easy to get it. To be a successful in a long run, one should try to understand in depth. We cannot have depth understanding of all, so one may scan quickly to get surface knowledge of the field. Once you know your topic of interest, sufficient effort has to be made utilizing all available resources on the topics of your interest.

3. Could you please share your favorite research papers? 
MLN:   There are no such special papers I published that I can say it’s my favorite one so far. If you have particular question about any of the papers I have co-authored please contact me by email or other ways.

4. As a professor, do classes have personality? Any tips on dealing with students in a big class and then as a single individual? 
MLN:   Personality matters everywhere since it is the first thing people make perception on you. But after all, student will judge based on how knowledgeable you are how well you can explain. That’s why beginner has to study more before the class. For big class, a professor especially beginner, has to design the class in order to make the teaching/learning process effective before the class. Depending upon the topic of the day, the class has to be careful designed how to introduce the topic, how to interact with students, how to get the feedback and how to test. First thing to do in a new class is to win students trust. Allow them to ask, or ask them to make sure most of the students understand what you are trying to teach. Keep eye contacts with everyone so that students know that they are being watched. These things also come with experience.
5. How to make a class effective? Is there a defined structure to achieve maximum benefit from teacher/student contact? 
MLN:   In order to make a class effective I use the following strategy. Once I have the topic for next class, I first think as if I were a student of the class, pretending as an average level student in the class. After that I check the goal of the topic. And then I design the class where to raise the question/ where to give examples or demonstration, give some time for discussion or activity, which problems to solve or ask them to solve, etc. Class should not be monotonous. That is why ‘Lecture’ mode of teaching is not considered now as an effective way of teaching. My personal experience also supports it. Especially teaching science in ‘Lecture’ mode is the least effective method.
There are lots of effective teaching methods developed recently by physics education research groups. All of them are intended to engage the students in the class and learn by actively participating in the class through hands-on activities, working together, discussions, project works etc.  Our traditional lecture rooms are not suitable for utilizing most of these kinds of teaching methods.

6. Could you share some effective teaching techniques that result in intended learning. 
MLN:   I have been using an active learning method called SCALE-UP (Student Center Active Learning Environment for University Physics) in general physics courses. Classroom for this teaching method is completely different from traditional ones. The SCALE-UP classroom has round tables like in a restaurant for 9 students in each table and they are divided into 3 groups. The three students in the group work together or learn each other. Each topic is generally introduced by hands-on activity. They collectively perform the activities and draw conclusion what they learned. They can also discuss with other groups. They also work together in problem solving. The classroom is made technology rich with laptop for each group, whiteboards around the classroom so that they can discuss with others or present their works. Instructor has to monitor students’ progress and help them to bring into the track. They also perform lab in the same room correlating the subject they are learning that week. There are regular quizzes which force students to prepare at home and help for feedback. Quite often I use Interactive Lecture Demonstration (ILD) in that class which enhances conceptual understanding. In ILD, students not only observe the demonstration, they have to think, predict what would happen, observe what really happens and discuss on the results. This helps to understand if there is any misconception. From my evaluation survey tests, average gain of the conceptual understanding in the SCALE-UP mode is about double of the traditional lecture mode.

7. Could you please share independent study/ projects with your students. Or any interesting moments with students. 
MLN:   There are many students I have mentored in independent study and research projects. They are mostly physics majors. Some students decided to go for graduate programs due to their involvement in the research projects.
I think my interesting moment with students would be in general physics class. There are always some students in the class who never had physics before. In their high school, they did not learn physics at all or do not know what happens in physics class. After a while they say, this course needs thinking and this professor really makes you think to understand in this class.

8. What have you found to be the roughest aspect of discipline, if any? 
MLN:   I think we are still not able to present physics in a simple way that everyone likes to know. Another rough aspect could be, although most people in science make significant investment of effort and life, their financial status in their life is mostly poor.

9. Tenure process is regarded as a daunting exercise. Could you please share tips on balancing life and work? 
MLN:   Tenure period is the time to prove you as a successful person in teaching, research and college service. Evaluation is done in most places based on these three categories. You have to show your significant contributions in all these sectors. Most institutes give more preference in research. However, poor teaching evaluation may also lead to denying from tenured. You have to check with your college and know how the evaluations are done.  Actually one has to spend more time on preparation for teaching in the beginning because you do not want to be a band professor in the class. In the mean time you have start thinking about research projects, writing proposals for grant, getting students in the lab for research projects. Some institutes evaluate every year for reappointment. One could be out before during the tenure track period. So you have to maintain the progress. Working on short term and long term projects in parallel is a useful idea. Some projects can be done in collaboration with others. Family support is a must. You cannot spend enough time (with them) until tenured. Family also needs to understand and cope with you for your success. There was a joke told in the orientation that tenure track professor does not see daylight until tenured.
Enjoying with the work including dealing with problems is the key to success. Once we enjoy with our work, it will be part of our life. Our life is successful if we are successful at work.

10. Sir, not being a professional interviewer, are there questions I have not asked that you wish I would have? Or anything to add? 
MLN:   There are lots of Nepalese physicists now in the USA working as post-doc. If anybody likes to discuss personally please do not hesitate to contact me. We have to increase this number as we did in the number of graduate students from Nepal. In order to get tenure track professor job, requirement has to be very well matched. Additionally, one has to compete with selected candidates in application process and interview process. One has to exhibit an outstanding quality to distinguish from others. So, it may need more pre-plans and quality works. If you are working as post-doc, my suggestions are, take opportunity of grant writing, teaching classes in undergraduate, mentoring students, designing your own research project and publishing articles.
Lastly, in order to improve quality of our work, we have to initiate research work while studying in Nepal. As a contribution to our motherland, I am proposing to establish a research fund to initiate physics research in Nepal. The fund will be used to support research projects in Nepal which can expand the research works and publish in international journals so that proposals can be developed for bigger grants. Contributions to the fund are collected from all physicists in the USA including graduate students. I think our small contribution every year will make a big change in our society.
Thank you Nabin ji.

NKM: Thank you for your time!  

This is an attempt to share successful Nepali physicists. The hope is to invite one personality every month. Please kindly suggest whom to invite next. 


-->

Wednesday, January 1, 2014

Presentations for 94th American Meteorological Society Annual Meeting Atlanta, GA


Monday, 3 February 2014: 11:15 AM

Regional estimates of ground level Aerosol using Satellite Remote Sensing and Machine-Learning
Room C204 (The Georgia World Congress Center )
Nabin Malakar, City College of New York, New York, NY; and A. Atia, B. Gross, F. Moshary, S. Ahmed, and D. Lary
The ground-level aerosols are known to have harmful impact on people's health. The Moderate Imaging resolution Spectroradiometer (MODIS) sensors onboard aqua and terra satellites retrieve aerosol optical depth (AOD) at various bands. The comparison between the AOD measured from the satellite MODIS instruments and the ground-based Aerosol Robotic Network (AERONET) system at 550 nm shows that there is a bias between the two data products. In this study we explore the factors that can delineate these extrema, and/or explain them statistically. We use the MODIS 3 km and 10 km resolution AOD products, and develop a machine-learning framework to compare the Aqua and Terra MODIS-retrieved AODs with the ground- based AERONET observations. The analysis uses several measured variables such as the MODIS AOD, surface type, land use, etc. as input in order to train a neural network in regression mode with a special emphasis on biases observed over non vegetative urban surfaces. The result is the estimator of the bias-corrected estimates of AOD. This research is part of our goal to provide air quality information, with special focus on the northeast region of the USA, which can also be useful for developing regional-level decision support tools.

Tuesday, 4 February 2014: 4:00 PM
A Regional NN estimator of PM2.5 using satellite AOD and WRF meteorology measurements
Room C206 (The Georgia World Congress Center )
Lina Cordero, City College of New York, New York, NY; and N. Malakar, D. Vidal, R. Latto, B. Gross, F. Moshary, and S. Ahmed
Besides affecting the global energy balance, aerosols can have a significant health impact. In particular, extended exposure ultrafine particles is a major concern and regulations by the EPA are constituted to deal with this issue. Unfortunately, measuring surface aerosols over wide areas is costly and difficult so the potential of using satellite remote sensing and/or models becomes an important area of study. In this presentation, we explore the potential of combining meteorological data together with column integrated AOD within a Neural Network approach. To begin, the study is isolated to New York City where accurate AERONET AOD as well as Lidar derived PBL heights along with weather station meteorology is included. The main result of this isolated study illustrates that beyond AOD, the next important factor is the PBL height. This result motivates an extended study where MODIS mosaic AOD's are combined with WRF weather forecast model inputs including PBL height. To use WRF PBL, a matchup between WRF and Calipso is given for single layer cases illustrating strong correlations in spring and summer when PM25 is most important. In particular, we find that with seasonal training, we are able to generally improve on the existing approach utilized by the IDEA (Infusing satellite Data into Environmental air quality Applications) product which utilizes MODIS AOD and GEOS-CHEM PM25/AOD factors. In addition, we explore potential improvements that can occur if we can filter aloft plumes from the processing stream using the NAAPS air forecast model as well as the use of EOF's to fill missing gaps in the AOD spatial imagery.

Thursday, 6 February 2014: 9:00 AM
Use of NN based approaches to create high resolution climate meteorological forecasts
Room C101 (The Georgia World Congress Center )
Nabin Malakar, City College, New York, NY; and B. Gross, J. E. Gonzalez, P. Yang, and F. Moshary
The effects of global climate forecasts on regional scale domains requires that the low resolution GCM forecast data can be intelligently modified so that it can be injected into high resolution models such as terrestrial ecosystems etc. This is often called downscaling in the climate forecast literature and is usually performed using one of 2 different strategies. In the first strategy, the use of purely statistical approaches such as interpolation is applied to the GCM low resolution data to provide the high resolution data. Of course, the “high” resolution data really does not possess any high resolution inputs that can drive regional scale models. In particular, valuable high resolution information such as land surface identification and potential emission sources is not used. On the other hand, the potential of using regional Meteorological Models such as WRF can be attempted where the GCM conditions and the forecasted land surface properties are encoded into a future time slice. Of course, this approach is extremely computer intensive and the performance may not be worth the computer resources. In this presentation, we make use of another intermediate approach where low resolution meteorological data including both surface and column integrated parameters are combined with high resolution land surface classification parameters within a NN training scheme in an attempt to improve on purely interpolative approaches. In particular, our study region is the North East domain [{35N,45N} x {-85W,-65W}] . In particular, we focus on High and Low temperature extremes which are the outputs to be considered are obtained within the PRISM data set while the low resolution climatology parameters at low resolution (.5 deg) MET data including Tmax, Tmin, Rhum, Wind Speed, Radiation, Precip and Planetary Boundary Layer height are obtained from the ISI-MIP climatology forecast database. In addition, a high resolution land surface map is used based on the 2006 USGS land surface map. Preliminary results show that the NN approach can result in improved high resolution performance in areas where land surface features change rapidly. In addition, we will make comparisons using the WRF model for the time periods from 2006-2011.

-->