System-Level Design Methods for Deep Learning on Heterogeneous Architectures
November 6, 2020
November 6, 2020
System-Level Design Methods for Deep Learning on Heterogeneous Architectures (SLOHA 2021)
co-located with Conference on Design, Automation and Test in Europe (DATE)
February 5, 2021
Submissions Due: November 6, 2020
*Scope of DATE Friday Workshop*
The SLOHA 2021 Virtual Workshop aims at providing a thorough overview of the most recent advancements in the topics related to enabling, optimizing and using the powerful deep learning (DL) technology on embedded systems, ranging from dedicated accelerator architectures to automated topology selection algorithms and use cases, through specialized compilers and optimized scheduling algorithms. The workshop will also be an excellent opportunity for networking in this emerging interdisciplinary field.
The main theme of the SLOHA 2021 Virtual Workshop will be system-level design methods for deep learning on heterogeneous architectures.
Topics of the SLOHA Workshop include, but are not limited to
- System-level design methods (e.g., scheduling, mapping) for DL
- Compilers, code generators, and synthesis tools for deployment
- Heterogeneous computing platforms, edge devices, neuromorphic hardware for DL
- Modeling of DL target platforms
- Design space exploration and optimization (accuracy, performance, energy efficiency, robustness,
- security and other non-functional properties)
- Deep compression, quantization and pruning techniques for edge devices
- Modeling of hardware-agnostic neural networks
- Signal processing applications, use cases, and best practices
*Guide for Authors*
We will accept two contributions formats: regular papers reporting original contributions (up to six pages) and extended abstracts describing works-in-progress (WiP, up to two pages).
All contributions should follow the standard IEEE conference template (http://www.ieee.org/conferences_events/conferences/publishing/templates.html); papers may or may not report authors names and affiliations (optional blind review). All papers must be submitted in PDF format via the conference management software EasyChair (https://easychair.org/conferences/?conf=sloha2021); additional details about the submission process are available on the workshop’s web page.
Accepted papers will be included in an ePrint collection with Open Access. Every accepted paper must have at least one author registered to the workshop by the time the camera-ready paper is due. The regular papers judged as most valuable will be allocated a slot for an oral presentation. Registered authors will be able to present their work during a virtual poster session; due to the virtual format of the conference, authors will need to prepare short presentations.
Submission deadline (regular papers): Friday, November 6, 2020
Notification of acceptance (regular papers): Wednesday, November 25, 2020