To Build the Next-Gen Intelligent Distributed Systems
I joined the Computer Science Engineering Division of Louisiana State University (LSU) as an Assistant Professor 2021 spring. I am looking for Ph.D. students who are self-motivated and have a strong interest in Machine Learning and Distributed Computing Systems. For more information about me and my research, please check my homepage.
I would like to share with you my story about bridging machine learning and distributed computing systems. If you happen to be interested, please drop me an email with your CV and transcript at [email protected].
The prosperity of Big Data and Machine Learning (ML) owes to the recent advances in distributed computing systems, such as Apache Spark and TensorFlow, which orchestrate numerous servers to process petabytes of data.
With components across datacenters, physical servers, virtual machines, sub-systems, and software stacks, distributed computing systems are like the natural ecosystems—dynamic, complex, and random—a bug flapping its wings might cancel your Uber Eats order.
In my Ph.D. study, I have been actively applying innovative machine learning techniques to understand the dynamic performance of distributed computing systems and build new systems with intelligent scheduling algorithms. We want to enable an autopilot in distributed computing systems.
It is of great FUN to bridge ML and distributed systems! By crafting innovative ML models, you will discover unknown behaviors and patterns inside the distributed architecture—processes, containers, servers, network links, and datacenters. Such discoveries eventually inspire you to rethink and redesign the distributed computing systems.
Yet, innovation just begins in this cross-disciplinary area.
IntelliSys Lab is committed to building the next-generation intelligent distributed computing systems with the following highlights:
Smarter learning algorithms that analyze and understand computation tasks such as CNN models, resource provisionings, and the performance of distributed systems.
Faster software and hardware architectures that fuse intelligent scheduling algorithms such as Reinforcement Learning, computing accelerators, and cloud computing.
Lightweight system design oriented to mobile devices and IoT sensors. We are pushing forward computational intelligence to devices closer to data and users.
We envision that you would be a good team player fitting into our research context with at least one of the following skills:
- A solid understanding of statistics, random process, and optimization theory. Students with no computer science background are also welcomed.
- Familiar with UNIX-like systems and the toolchains of cloud computing. A solid understanding of operating systems, computer networks, and at least one programming language (e.g., Python, C/C++, and Java).
- Knowledge in ML theory on deep learning and reinforcement learning, and experience with frameworks such as TensorFlow, PyTorch, and MXNet.
- Understand the internals of distributed computing systems (e.g., Apache Hadoop and Spark), and programming experience with CUDA.
Please note that we can relax the requirements of GRE, TOEFL, and IELTS scores if you match our expectations.
First of all, this area is exciting and promising that both academia and industry are seeking talents with expertise in Machine Learning and Distributed Systems.
You won't be working alone, and we will closely work together. As your supervisor, I am very willing to share with you my experience of seeking both problems and solutions. Ph.D. is hard, but it would be much less suffering if you work with the right people in the right place.
With a very high reputation in research, LSU is the flagship R1 university and the most beloved university in the state of Louisiana. Our scholarship will cover your tuition fee and expenses to live a decent life here.
Again, if you are interested in joining us, please drop me an email at [email protected].