Number of Positions: 3

Extent:     Full-time

Duration:     temporary

Application deadline: 17.06.2024

Beginning:     as soon as possible

Further Information

The application deadline is September 15, 2024. Please send your application (CV and grade sheet) directly to the potential supervisor.

The run-time reconfigurability and high parallelism offered by FPGAs make them an attractive choice for implementing hardware accelerators for ML algorithms. In the quest for designing efficient FPGA-based hardware accelerators for ML algorithms, the inherent error-resilience of ML algorithms can be exploited to implement approximate hardware accelerators to trade the output accuracy with better overall performance. As multiplication and addition are the two main arithmetic operations in ML algorithms, most state-of-the-art approximate accelerators have considered approximate architectures for these operations. However, these works have mainly considered the exploration and selection of approximate operators from an existing set of operators. To this end, this project focuses on designing a reinforcement learning (RL)-based framework for synthesizing and implementing novel approximate operators. RL is a type of machine learning where an agent learns to perform actions in an environment to maximize a reward signal. RL-based techniques would help achieve approximate operators with better accuracy-performance trade-offs in this project.

• Pre-requisites:
  • Digital Design, FPGA-based accelerator design
  • Python, TCL
  • Some knowledge of ML algorithms
• Skills that will be acquired during project work:
  • ML for EDA
  • Multi-objective optimization of hardware accelerators.
  • Technical writing for research publications.
• Related Publications:
  • S. Ullah, S. S. Sahoo, and A. Kumar. "CoOAx: Correlation-aware Synthesis of FPGA-based Approximate Operators." Proceedings of the Great Lakes Symposium on VLSI 2023. 2023.
  • S. Ullah, S. S. Sahoo, N. Ahmed, D. Chaudhury, and A. Kumar "AppAxO: Designing App lication-specific Approximate Operators for FPGA-based Embedded Systems." ACM Transactions on Embedded Computing Systems (TECS) 21.3 (2022): 1-31.
  • S. Ullah, S. S. Sahoo, A. Kumar, "CLAppED: A Design Framework for Implementing Cross-Layer Approximation in FPGA-based Embedded Systems", In Proceeding: 2021 58th ACM/IEEE Design Automation Conference (DAC), pp. 1-6, Jul 2021.
• Contact: Salim Ullah

Compared to ASIC and FPGA, CGRAs is a more viable hardware platform for implementing applications in IoT edge devices, due to their promising trade-off in performance and energy-efficiency. In our recent papers (see one here), we have designed an approximate CGRA which can execute various applications from biomedical to image/video processing. For the follow-up journal paper, we want to extend the mapping to support for 5G applications. So we are looking for a 3-6 month student assistant, who will be a co-author in this hot topic.

• Tasks:
  • Generate the Data Flow Graph (DFG) of applications using e.g., LLVM compiler
  • Mapping applications' DFG using Morpher/OpenCGRA mapper
• Requirements:
  • Verilog/VHDL
  • Experience with ASIC mapping tools (Morpher, OpenCGRA, CGRA-ME, etc)
  • DFG generation using LLVM
  • Installation of open-source (github) tools on Ubuntu
• Contact information for more details

Zahra Ebrahimi

In general, there are three categories of ML techniques -- supervised-learning, unsupervised-learning, and reinforcement-learning -- where depending on the problem, parameters, and inputs, only some of these techniques are suitable and used for system properties optimization. These ML techniques are memory-intensive and computationally expensive, which makes some of them incompatible with real-time system design due to the overheads, which may cause an effect on applications' timeliness. Therefore, this project aims to analyze and investigate various ML techniques in terms of overheads, accuracy, and capability and determine the efficient ones suitable for embedded real-time systems.

• Pre-Requisites
  • Proficiency in C++, Python, Matlab
  • Knowledge about Machine Learning techniques
  • Good knowledge of computer architecture and algorithm design
• Related Publications:
  • S. Pagani, P. D. S. Manoj, A. Jantsch and J. Henkel, "Machine Learning for Power, Energy, and Thermal Management on Multicore Processors: A Survey," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), vol. 39, no. 1, pp. 101-116, 2020.
• Contact
  • Behnaz Ranjbar