Lecturer
- 51cg
-
- School/Department
- School of Natural and Computing Sciences


Biography
I'm a Lecturer (Assistant Professor) in the Computer Science Department at the 51cg. My research focuses on program optimisation and large-scale DNN (Deep Neural Network) model acceleration towards heterogeneous platforms (CPU, GPU, TPU, FPGA, etc.).
Before joining the 51cg, I was a postdoc research fellow at Trinity College Dublin and the University of Edinburgh, where I received my PhD in computer science.
I'm looking for highly motivated PhD students interested in compiler techniques, machine learning and heterogeneous systems. Please feel free to contact me for any enquiries.
- Research
-
Research Overview
My research interests include:
- Hardware/Software co-design approaches to efficiently deploy Deep Learning applications on mobile, edge and heterogeneous devices.
- Acceleration of deep learning models via compilation techniques
- AI compilers
- Heterogeneous computing
- Efficient FPGA code generation
Research Areas
Accepting PhDs
I am currently accepting PhDs in Computing Science.
Please get in touch if you would like to discuss your research ideas further.
- Teaching
-
Teaching Responsibilities
- JC1001 Python Programming Foundation
- JC2503 Web Application Development
- Publications
-
Page 1 of 1 Results 1 to 3 of 3
From Threads to Tiles: T2T, a Compiler for CUDA-to-NPU Translation via 2D Vectorization
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] DOI:
ParDiff: Efficiently Parallelizing Reverse-Mode Automatic Differentiation with Direct Indexing
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] DOI:
- [OPEN ACCESS]
SpaceServe: Spatial Multiplexing of Complementary Encoders and Decoders for Multimodal LLMs
Conference on Neural Information Processing SystemsContributions to Conferences: Posters- [ONLINE]
- [ONLINE]
