Many companies, such as Google, Facebook, and Amazon are building new specialized programming frameworks. This is because these companies need to allow their users to write simple, high-level code and run it efficiently on different hardware architectures. For example, Google has built TensorFlow, a framework for deep learning allowing users to run deep learning on multiple hardware architectures without changing the code.
Our research team at NYUAD (New York University Abu Dhabi) in collaboration with MIT (Massachusetts Institute of Technology, USA), Google, and Adobe, is developing a new programming framework called Tiramisu [1]. Unlike existing frameworks, Tiramisu can perform advanced code optimizations that are hard to apply otherwise. Because of this, Tiramisu can generate fast code that outperforms highly optimized code written by expert programmers and can target different hardware architectures (multicore, GPUs, FPGAs, and distributed machines). Three projects related to Tiramisu are open to students.
Automatic Code Optimization Using Deep Learning
In order to have the best performance (fastest execution) for a given Tiramisu program, many code optimizations should be applied. Optimizations include vectorization (using hardware vector instructions), parallelization (running loop iterations in parallel), enhancing data locality by fusion, and blocking (i.e. accessing arrays in a way that improves temporal and spatial data locality). A large number of optimizations exist and choosing which optimization should be used and which should not is important for performance. In some situations, some optimizations are harmful to performance while they are beneficial in other situations. Currently, there is no way to help users choose which optimizations should be used. Expert Google programmers usually spend a lot of time trying different optimizations manually to find the best set of optimizations. The goal of this project is to add support for automatic code optimization in Tiramisu. In particular, we want to use machine learning/deep learning to achieve this. Currently, a basic automatic optimization module that relies on machine learning has been developed and we want to take that module to the next level. The final product of this project would be a compiler pass that allows Tiramisu to automatically choose which optimization should be used for a given unoptimized program. We want to produce a high-quality technique that can be used by the users of Tiramisu and especially by our partner companies and research labs.
Benefits
The selected students will work on cutting edge projects with a very high practical impact on the areas of deep learning and nuclear physics (we are integrating Tiramisu in PyTorch, and in the USQCD frameworks, therefore, all future users of these frameworks will benefit from your contribution).
The selected students will spend 3 months at NYUAD in Abu Dhabi and will receive a monthly salary of 2900$. Travel and VISA expenses will also be paid by NYUAD. The students will participate in writing papers and submit them to the best conferences in the field (NIPS, PLDI, ISCA, ...). The projects will include close collaboration with researchers from MIT and Harvard. Regular Skype calls will be organized with MIT and Harvard researchers.
Application
To apply, please send your CV and transcripts (with ranking) for the last two years and indicate when you'll be able to start. Being able to start as early as possible (after finishing your classes) is a plus. Please indicate which projects are interesting to you (it is Ok to express interest in all of them). Please also indicate any relevant technical skills that you have. Do not hesitate to ask questions by email: baghdadi@nyu.edu.
NYU Abu Dhabi is a degree-granting research university with a fully integrated liberal arts and science undergraduate program in the Arts, Sciences, Social Sciences, Humanities, and Engineering. NYU Abu Dhabi, NYU New York, and NYU Shanghai, form the backbone of NYU's global network university, an interconnected network of portal campuses and academic centers across six continents that enable seamless international mobility of students and faculty in their pursuit of academic and scholarly activity. This global university represents a transformative shift in higher education, one in which the intellectual and creative endeavors of academia are shaped and examined through an international and multicultural perspective. As a major intellectual hub at the crossroads of the Arab world, NYUAD serves as a center for scholarly thought, advanced research, knowledge creation, and sharing, through its academic, research, and creative activities.
UAE Nationals are encouraged to apply.
This position is not located in the United States. You must be willing to relocate to Abu Dhabi, United Arab Emirates.
For people in the EU, click here for information on your privacy rights under GDPR: www.nyu.edu/it/gdpr
NYU is an equal opportunity employer committed to equity, diversity, and social inclusion.
Internal Number: 79463
About New York University
Founded in 1831, New York University is now one of the largest private universities in the United States. Of the more than 3,000 colleges and universities in America, New York University is one of only 60 member institutions of the distinguished Association of American Universities.
