MIT-Toyohashi ASPIRE workshop 2025 on June 26th at Science Tokyo.
Date: 26 June, 2025
Location: Science Tokyo Front, Ookayama Campus, Institute of Science Tokyo (Kuramae-Kaikan、蔵前会館 2F 大会議室)
Program
**All programs are in JST (UTC+9)
10:00 JST ASPIRE project introduction (Yukinori Sato, Toyohashi Univ. of Tech.)
10:15 Saman Amarasinghe (MIT): Compiler 2.0: Building Compilers in the Era of Machine Learning
11:15 Ajay Brahmakshatriya (MIT): Democratizing High-performance DSL Development with the BuildIt Framework.
12:15 Lunch Break
12:45 Poster presentations by students
13:45 Hayato Yamaki (UEC): High-Speed Packet Compression and Forwarding System.
14:15 Yudai Tanabe (Science Tokyo): Bringing Fine-Grained Task Parallelism to GPUs
14:45 Break
15:00 Katsumi Okuda (Mitsubishi Electric): Programming with LLMs: From Unified Interfaces to Performance Optimization
15:30 Ryuichi Sakamoto (Science Tokyo): Accelerating microservice using SmartNICs
16:00 Yukinori Sato (Toyohashi Univ. of Tech.): DSL-based automatic code optimization for AI accelerators and SmartNICs
16:30 close
We prepare Zoom in the morning sessions (10:00am – 12:15pm). To participate in the workshop via Zoom, please make a registration in the following form.
Registration
Please make a registration from the following URL:
https://forms.gle/1co46Xc1iXaA2bHg9
A part of this tutorial will be supported by JST ASPIRE, Grant Number JPMJAP2430.
Project overview (in Japanese)
Compiler 2.0: Building Compilers in the Era of Machine Learning
Saman Amarasinghe (MIT)
Abstract: Modern compilers are among the most critical, complex, and widely used software systems—yet they are still largely built on decades-old technologies. The rise of machine learning is fundamentally reshaping what programming means. It not only has the potential to revolutionize how we build compilers but may even render traditional compilers obsolete. In this talk, I will explore the current impact of machine learning on compilers, discuss near-term opportunities, and offer a few predictions about what the future may hold.
Saman Amarasinghe is the Thomas and Gerd Perkins Professor in the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL), where he leads the Commit compiler group. Under his leadership, the Commit group has developed a wide range of innovative programming languages and compilers, including StreamIt, StreamJIT, PetaBricks, Halide, TACO, Finch, SySTeC, GraphIt, Simit, MILK, Cimple, BioStream, NetBlocks, BREeze, CoLa, Shim, AskIt, and Seq. Additionally, the group has created compiler and runtime frameworks such as DynamoRIO, Helium, Tiramisu, Codon, BuildIt, and D2X as well as tools for vectorization like Superword Level Parallelism (SLP), goSLP, and VeGen. Saman’s team also developed Ithemal, a machine-learning-based performance predictor, Program Shepherding to protect programs from external attacks, the OpenTuner extendable autotuner, and the Kendo deterministic execution system. He was also co-leader of the Raw architecture project. Outside academia, Saman has co-founded several companies, including Determina, Lanka Internet Services Ltd., Venti Technologies, DataCebo, and Exaloop. He earned his BS in Electrical Engineering and Computer Science from Cornell University in 1988, and his MSEE and Ph.D. from Stanford University in 1990 and 1997, respectively. He is also a Fellow of the ACM.
Map
Ookayama Campus, Institute of Science Tokyo
Conference Room L (2nd Floor), Science Tokyo Front
(Kuramae-Kaikan、蔵前会館 2F 大会議室)