Kareem H
- Research Program Mentor
PhD candidate at Stanford University
Expertise
Physics, Computer Science, Machine Learning, Math, Data Science
Bio
I am Physics PhD student at Stanford University with a B.S. in Physics and Math from the University of Michigan. I started my academic career I working at CERN where I constructed part of the particle detector and contributed to the Higgs Boson (God Particle) discovery through various analyses. I now work at SLAC national lab as I work towards my PhD, here I use ultrashort relativistic electrons to image basic chemical processes in order to better understand many-body quantum mechanics. One of my projects uses machine learning, where work in collaboration with Google Brain. In my spare time I do machine learning research with for Google X. My teaching experiences ranges from one-on-one to teaching 100 kids at a time. In addition to teaching college students as a teaching assistant I have mentored a handful of high school students in the past and I teach a 2 hour 4th grade level CS to hundreds of kids online. I as well volunteer at local high schools helping under-resourced students do science fair projects and apply to college.Project ideas
Modeling a Galaxy and the Effects of Dark Matter
- You will learn about gravitational physics and some basic general relativity. - We will go over why dark matter must exist, the evidence for it, and the possible particles that may make it up. - You will simulate a spiral galaxy and be able to calculate how many stars and how much matter should exist - You will model this galaxy with and without dark matter - You will reproduce the evidence you learned about as to why dark matter must exist - You will change the properties of the dark matter to see what galaxies would look like for different types of dark matter or with your own made up dark matter particles.
Artificial Intelligence and Machine Learning
- Machine learning/artificial intelligence is a wide field and we can do anything you want, if you have a problem you've ever been curious we can try and solve it with machine learning. - Do you have a favorite computer game? We could train a machine to play it. - Do you want a computer to answer your homework questions for your? We can make it do that. - Do you want to turn your pictures into artwork that looks like your favorite artist (or combination of artists) painted? We can do that. - You will also learn the fundamentals of machine learning from linear regression to neural networks and bayesian inference learning. - You will learn the math behind the algorithms and the tools to quickly build up advanced neural networks in a small amount of code: tensorflow, pytorch, etc.
Using Machine Learning to Predict Cryptocurrency Prices or Stock Prices
- I have a personal interest in this project and I have been collecting cryptocurrency data and I have built a back-testing setup from which you can test your predictive algorithm to see how profitable it is. - If the student is interested in cryptocurrency and blockchain technology then we will go over what the blockchain is, how does it enforce security, exactly how bitcoin works, as well as the threats that could break it and how they could be fixed. - We will cover statistics, general techniques for reading price history of stocks and cryptocurrencies to gain a sense of where the market is, and some indicators that help to predict it. - You will also learning everything stated in the above project in addition to some more focused sequential based analyses, like natural language processing. - You will also create your own predictive algorithms and test them with historical data, if they work well enough we could even test them on the live market with small funds I could provide. - You will learn how to program in python and a new machine learning language being developed at Google that is beginning to replace the language that is currently most popular for machine learning.
Molecular Dynamics Simulation
- This project offers a lot of different areas to study - You will build a force field molecular dynamics simulation that won the 2013 Nobel Prize in chemistry - You will learn how to program in either python or C++, by the end of this project you will be proficient enough at programming that you could likely skip the college introductory programming class. - You will learn about fundamental physics principles like Newtonian mechanics, equations of motions, the Lagrangian, electrodynamics, and we can add thermodynamic calculations if this is also an interest. - If you are interested we can cover some basic quantum mechanics. - This project can also have a machine learning aspect to it. The algorithms that won the Nobel Prize could be considered basic machine learning and we could improve them using more advanced machine learning techniques. - Given the breadth this project can encompass we will talk about your interests and design a project best suited to them.
Building Quantum Computer Algorithms
- By the end of this project you will build a quantum computer algorithm and run it on a real quantum computer. - Before you can build this algorithm you will first have to understand how quantum computers work. - You will learn a lot of basic quantum mechanics, enough to carry you through a third of an undergraduate quantum course. - You will learn about quantum entanglement and how this makes quantum computers so powerful. - You will learn about how quantum computers may be built in the future (this is still an open question in physics). - You will learn the quantum gates that are needed to create any possible quantum algorithm. - You will learn some famous quantum algorithms and pick one of them to program and run on a quantum computer.
Coding skills
Python, C++, Bash, TensorflowTeaching experience
- One on one tutoring for high school math - Mentoring for a high school science fair project - Teaching college level classes - Teaching large 4th grade CS coursesCredentials
Work experience
Education
Reviews
"It was very educating. Though I didn't create as many things as I'd have liked, I went from having no knowledge about NLP in the beginning of summer to having a good understanding of all the NLP algorithms even up to more recent ones."
