Attention ML compiler enthusiasts! Join us for the CODAI’23 workshop where we’ll discuss AI compilers and accelerators for Edge AI. This workshop brings together academic and industrial researchers to collaborate on deploying neural networks on edge devices. We’ll cover topics such as compilers, optimization techniques, code-generation, hardware-backends, and more.
More information: CODAI Workshop
Last year there were excellent contributions also from the TVM community:
- @tqchen: Keynote: Abstract for Machine Learning Compilations
- @areusch: Whole-model optimization with Apache TVM
- @PhilippvK, @r.stahl et al: MLonMCU: TinyML Benchmarking with Fast Retargeting
- Federico Nicolás Peccia et al: Integration of a systolic array based hardware accelerator into a DNN operator auto-tuning framework
- and many more
We are looking forward to contributions from members of the TVM community this year, as well!
Call for Papers! CODAI Workshop, co-hosted at Embedded Systems Week
September 17th - 22nd September 2023
Hamburg, Germany (Hybrid)
Abstract Submission Deadline: July 6th 2023
Paper Submission Deadline: July 13th 2023
The CODAI workshop is a great opportunity to share and learn about the latest developments of Edge and Embedded AI in the academia and industry.
The topics of interest include, but are not limited to:
- Compilers for Edge AI: partitioning, µC, heterogeneous systems, intermediate representations or (domain specific) languages
- Optimization techniques and performance estimation of neural networks on edge device: e.g., compression, quantization techniques, virtual prototyping
- Code-generation and hardware-backends for Embedded AI accelerators - especially for RISC-V is appreciated
- Applications: processing of embedded vision, time-series data, etc.
- Novel brain-inspired algorithm for Edge AI
- Compiler and optimization techniques for beyond-von-Neumann AI accelerators