**start. **I’m currently taking a quantum computing class at pace university. this is a master/doctoral course and i’m an undergrad in his senior year. i was inspired to take this class from my visit to cern in geneva switzerland.

i’ll admit, i’m a bit of a dreamer. but i took this away from the experience: *think in leaps*

i’ve been reading the nielson and chuang book, the rieffel was a little rough to start out with. through perseverance and a weird connection with my experience taking organic chemistry, i was able to obtain the highest grade in the class! this, along with validation (it’s powerful), led me to spend most of my spring break reading…..

**we were tasked with doing a project.** i’ve been tracing the classic quantum algorithms but could not wrap my head around *what to do*.

reading led me to this, *solving the optimal trading trajectory problem using a quantum annealer*. these researchers, developed a multi-portfolio optimization algorithm using d-wave’s qcomputer. they were inspired by this article.

**my take.** these researchers did this on a d-wave qcomputer. that is a qcomputer specialized for quantum annealing. to forward my understanding of translating classical problems into quantum problems. i think it would be useful to work off these two articles (/&background) and create one for ibm’s universal qcomputer. that is:* an algorithm was designed for a computer that is specialized for one thing, what would happen if this was run on a generic universal qcomputer*? this would also provide me guided reference while also challenging myself. i also think it would be interesting to see how the two computer’s compete!

useful links:

algassert nielsonsolutions sarronson testyourself

end for today. march 27, 2019

sound for this section on soundcloud

*addendum* *1*

**more details.** ok so it’s been tough since the last post! jon and i got started on writing our paper. this will be a full blown ieee formatted project paper. we designed the paper to talk about translating classical -> quantum implementations. specifically, we wanted to focus on our troubles, what concepts could be made simpler and the culmination: the portfolio optimization. i took out a book in the library on optimizations to understand what it all meant. to put it into one sentence, we calculate the portfolio using either a variational quantum eigensolver or a quantum approximate optimization algorithm. we then compare our results by measure of sharpe’s ratio. this metric gives us an idea of the desirability of each portfolio. ** **

**hiccups. **what makes this problem wonderfully hard is that we needed to understand what was going on in the classic case. then, apply that to our qcomputer. it’s been hard finding the resources on how to find real problems as well as translations. for the first problem, instead of generating random matrices that are hard to understand, we now have real values to work with (above table). also, luckily, we found a bunch of resources that will hopefully give us more practice. one big concern for us is if we have enough qubits… *time will tell.* I’m excited!

(qiskitpo, qtsp, po)

**we visited ibm!** last week, our class went to their research center in yorktown. we got to see a *real *quantum processor!!! it was a great day of lectures on qc, blockchain, ai. there were also ibmrs? whose post’s i read on medium- it felt like a celebrity event. i think it was interesting to see how motivated people are to *create*.

useful links:

qalgoblog everwonderwhataseedwas goodre-intro

end for today. april 8, 2019

sound for this section on soundcloud

*addendum 2.*

**we changed gears**! ok so i’m back after publishing our project in the 17th annual michael gargano research day and presented on may 3rd. writing a formal paper was another great experience. the financial quantum computing was changed because we didn’t have enough qubits to work with! instead we worked on solving the traveling salesman problem; which was the original problem. we used qiskit to implement it using the variational quantum eigensolver (ry variational form) and the spsa optimizer. i think it took two main things away from this experience. 1. quantum computing is still limited, we could only do “toy problems.” these were problems that are easily solvable on classical computers but are used as a proof of concept for quantum computers. 2. qiskit abstracts away *a lot* of the physical implementation with their aqua package. this is good for someone who wants to just implement. however, we found ourselves relying on papers to see how it actually worked! this is probably the last addendum for this blog- on to another project!

thanks for reading 🙂

end for today. may 5, 2019

research page

## What do you think?