Research
My research interests span computational science, educational effectiveness, human-centered design, and equivalent access for people with disabilities. I have a track record in a traditional academic publishing environment and continue to present at conferences. I also advise students who publish and participate in standard scientific research groups sporadically. However, I have not been concerned with publishing traditional scientific research for the last few years as I have been focusing mainly on teaching and software development.
Effectiveness of computer programming to bolster STEM learning
[In progress at MIT]
In my role as computational curriculum advisor in the Department of Materials Science and Engineering I have been consulting with other departments looking to incorporate computer programming into the core curriculum. We are collecting data to understand when and for whom the inclusion of specially designed computer programming exercises can improve learning outcomes in online STEM education. For examples of the type of interactive graphics we will be incorporating you can read this nice blog by Stephen Wolfram.
In my role as computational curriculum advisor in the Department of Materials Science and Engineering I have been consulting with other departments looking to incorporate computer programming into the core curriculum. We are collecting data to understand when and for whom the inclusion of specially designed computer programming exercises can improve learning outcomes in online STEM education. For examples of the type of interactive graphics we will be incorporating you can read this nice blog by Stephen Wolfram.
Pedagogical effectiveness of interactive STEM graphics
[In progress on MITx]
In my role as computational curriculum advisor in the Department of Materials Science and Engineering I have been tasked with the incorporation of interactive visualizations that are embedded inside of online classes. We are collecting data to understand when and for whom the inclusion of these components increase learning outcomes. For examples of the type of interactive graphics we will be embedding you can see the Wolfram Demonstrations Project.
In my role as computational curriculum advisor in the Department of Materials Science and Engineering I have been tasked with the incorporation of interactive visualizations that are embedded inside of online classes. We are collecting data to understand when and for whom the inclusion of these components increase learning outcomes. For examples of the type of interactive graphics we will be embedding you can see the Wolfram Demonstrations Project.
Accessibility and educational equivalency of interactive STEM graphics for people with disabilities
The DIAGRAM Center is the research branch of Benetech Inc. (Beneficent Technologies). I have been working with Benetech since 2013 when I was awarded a grant to compile a set of best practices for providing pedagogically-equivalent information to blind people about content within interactive scientific graphics. Since 2013, I have attended the biannual planning meetings and worked on the Developers Working Group.
In keeping with the DIAGRAM Center goal of dramatically changing the way image and graphic content for accessible instructional materials (AIM) is produced and accessed, DIAGRAM both conducts its own research and partners with other organizations to conduct research to achieve this goal. One such research report on Accessible Dynamic Scientific Graphics was completed by Kyle Keane of Wolfram Research in June 2014. This project surveys the current best practices for making dynamic scientific graphics accessible to persons with severe visual impairments in order to better understand the most effective practices for providing pedagogically-equivalent information about common scientific visualizations using audio feedback and verbal description.
I am quite pleased at the reception of my message that we need to start thinking about how to offer pedagogically-equivalent information to blind users of interactive graphics. My recommendations for a separation of intent from action spurred the creation of the User-Intent Working Group in the W3C. My guidelines for verbal description has been adopted by the team at University of Colorado who are building a library of interactive science teaching tools called PhET simulations. these simulations are the best examples of accessible interactive graphics for science education and I am deeply grateful for the chance to contribute to their development.
Inverse design of photonic crystals using an adaptive mesh search algorithm
In photonic crystals, infinite combinations of physical parameters can give rise to a rainbow of colors. Often, one wants to engineer a crystal with a given perceived color, but this inverse design problem requires finding a multitude of possible candidate parameters combinations. We built an adaptive mesh search algorithm that searches the continuous color space formed by parameter combinations for colors that match within the limits of human perception. This work makes it possible to bring structure color into the world of design since we can now predictably create stop signs, counterfeiting detection bands, and works of art without having to spend tons of resources with trial and error to discover the right engineering process. I am very proud that my student was able to publish this work in the Optical Society of America JOSA B.
Solid-state dewetting for thin metallic films
For a year, I worked as a research scientist in Carl V. Thompson's group at MIT. My goal was to simulate the process of solid-state dewetting that occurs when a thin metallic film is placed on top of another metal from which it wants to "dewet" (think of oil clumping on top of water). To solve the challenge, I developed a new method of atomistic kinetic Monte Carlo simulation for studying solid-state dewetting phenomena in thin metal films, extensible to any lattice structure. I transferred the techniques of big data and data science to develop algorithms that modeled the interactions of atoms on a lattice as if they were visitors to a website. Using modern computing techniques, I parallelized the algorithm using Bayesian statistics to efficiently use graphical processing units (GPUs) to perform massively parallel physical simulations. In the end, I was able to reproduce the phenomenology of the so-called "fingering instability" characterized by the asymmetric retraction of the edges of the film leaving behind long islands of material. I also reproduced the Rayleigh-like instability of a thin wire of metal breaking apart into small droplets.
Quantum error detection and correction for robust quantum computing with superconducting qubits
My PhD work involved designing and simulating currently realizable quantum error correction procedures for superconducting phase qubits. Although experiments in other superconducting systems have recently appeared, their main purpose is a proof-of-principle demonstration of quantum error correction; whereas my paper will look at procedures that can be implemented in a laboratory for real-world improvement of the storage capacity for quantum information. You can read the resulting paper in the American Physical Society's Physical Review A.
Applications of partial (reversible) quantum measurements
In quantum mechanics courses through the usual curriculum students are rarely taught about partial measurement. I do not believe this is a major tragedy since most of the time full measurement is performed. However, the lack of exposure to the subject of weak measurement seems to have left a gap in scientists understanding of the fundamentals of quantum mechanics. Also, since experiments rely on ensemble averages of multiple repeated projective measurements on similar operational procedures for tomographic verification, many people believe that the inherent randomness of quantum systems is the fundamental understanding. Converse to this assertion, quantum computing has shown us that the inner workings of the state description (wavefunction or density matrix) can have real effects on the final outcome of a single projective measurement after certain sequences of operations (quantum algorithms). The importance of these subtle differences in the philosophical interpretations of measurement's role in quantum mechanics may not at first seem important, but it is my belief that they allow us a deeper ability to engage in what I call quantum engineering (the use of the principles of quantum mechanics to achieve certain operational/computational goals, such as efficiently searching in an unsorted database or securely transmitting a cryptographic key). My work so far in Weak measurement has revolved around weak measurement reversal (also known as "uncollapsing").
As an introductory project when I began working with Dr. Korotkov I was asked to simulate a simple uncollapsing experiment that was performed by John Martinis' group at the University of California, Santa Barbara. While simulating the experiment I started to generalize the parameters and found a unexpected result: I could increase the fidelity of the procedure by increasing one of the non-operational durations relative to the others. This phenomena was soon easily understood as the reversal of the state-description dynamics caused by energy relaxation of the system. We realized that this was an effective method to suppress the decoherence caused by energy relaxation and could be implemented using two partial measurements of a single qubit. This was a very interesting result since the only other decoherence suppression technique that used only one qubit is Dynamical Decoupling, which cannot protect against Markovian processes (energy relaxation is a Markovian process). We published this result in American Physical Society's Physical Review A.
Early in 2011, an experimental group at POSTECH demonstrated our proposed procedure in a beautiful optical adaptation published in Nature (our original was mentioned in the context of superconducting qubits). Their results were outstandingly close to our theoretical predictions. After this accomplishment they extended the procedure to two entangled photons and showed that the procedure applied independently to each would in fact restore the entanglement lost to energy relaxation (this was expected and was mentioned as a footnote in our original paper). This is another very interesting effect, since even beyond the point of so-called entanglement sudden death (a premature disappearance of entanglement, even before full decoherence of the constituent subsystems), full entanglement can be restored to the photons.
Frustrated magnetism on the Kagome lattice
I first entered this topic with a very clear research question given to me by my advisor at the time, Dr. Kirill Shtengel: What is the ground state of 18 electrons on a Kagome lattice? Having sole responsibility over such a well-defined problem was a great joy for me. Dr. Shtengel was a wonderful mentor and gave a lot of his time introducing me to the conceptual techniques that we would later use in our search for a solution. Once I had an understanding of these techniques, it was off to the races. Well, it was off to the computer screen to start coding an algorithm to efficiently find the first stage of the solution. The task I performed was to find an array of all possible pairwise connections of 18 identical points equally spaced along the circumference of a circle. This was a really fun challenge and after considerable work I had a program that could do this task in a reasonable amount of time (it took about seven days continuous run time on my old laptop). The next stage was to turn this overcomplete basis into a orthonormal basis using a generalized Gramm-Schmidt procedure. This was a fairly straightforward program to build and I successfully completed this stage in the research. Although I found the topic stimulating and rewarding, I started to slip into a bit of a personal existential dilemma: I wanted to participate in the process of pure scientific inquiry, but I craved the stability of working toward a well-defined application. Over the years since this experience, I have come to believe that my primary purpose as a scientist is to maintain the chain of science, ethically employ my expertise in the use of funds given to me for this purpose, and to facilitate the progress of scientific discovery and understanding. However, as a young, idealistic developing scientist this desire for a stable purpose, and the discovery of Dr. Alexander Korotkov (a theoretical physicist working in the Electrical Engineering department), led me to my PhD work in Weak Measurements, Quantum Error Correction, Decoherence Suppression, and Quantum Communication.
Simulation of the thermodynamical self-assembly of virus capsids
The first research topic in gradute school that I approached was Biophysics. I had a long-held desire to understand the human experience from a physics-guided computational viewpoint and hoped this would be a path to finding that understanding. My work in Biophysics was under the supervision and guidance of Dr. Roya Zandi, although my primary collaborator was a postdoctoral researcher named Artem Levandovsky. Artem had developed a code that calculated stable virus formations built from elementary components and I worked with him for about three months coding a subroutine that would characterize and statistically analyze the population proportions of different stable configurations. Our aim was to show that the population distributions found by his code were consistent with the population variations found in nature.
Precise measurement of atomic and molecular differential scattering cross sections
While working toward my BS in physics, I was also a member of Dr. Murtadha Khakoo's electron scattering laboratory. I was involved in typical tasks such as data acquisition and analysis, equipment control and calibration, and on-the-fly troubleshooting and problem resolution. I also held a very special position as the only physics student with full access to the campus machine shop. After training with the campus's only employed machinist, David Parsons, I fabricated an entire vacuum-friendly rotary table for the angular study of differential electron cross sections scattered off of various polyatomic targets. This is the time in my life when I truly found a passion for physics as a philosophy of discovery and understanding of the natural world. To view my publications that resulted from my time in this lab, please see my Publications.