Research
My research explores the efficient realization of computation in silicon, combining innovations in algorithms, architectures, circuits, and EDA methodologies to design energy-efficient ASIC and FPGA computing systems.
Publications
AI and Machine Learning Accelerators
- Attention parallel hardware accelerators for Transformers and LLMs
- Spiking Neural Networks on digital hardware
- CNN dataflow hardware accelerators (Py2HLS)
Privacy-Preserving and Secure Computing
- Hardware accelerators for homomorphic-encryption-based privacy-preserving computation
- Customized Systolic arrays for accelerating FHE and PQC workloads
Processor and Computer Architecture
- Ultra low cost bit serial RISCV CPUs
- RISC-V Vector Processors
- DRIM: Energy efficient OoO RISCV processor dual issuing only compressed instructions
Awards and Honors
- Best Student Paper Award for my PhD students Christodoulos Peltekis and Kosmas Alexandridis at IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) 2024 for the paper "Reusing Softmax Hardware Unit for GELU Computation in Transformers"
- Best Student Paper Award for my PhD student D. Mangiras - Electronics, Circuits and Systems Track of Modern Circuits and Systems Technologies (MOCAST) conference for the paper "Incremental Lagrangian Relaxation based Discrete Gate Sizing and Threshold Voltage Assignment", July 2021.
- Best Paper Award - T track in Design Automation and Test in Europe conferene for the paper "Error-Shielded Register Renaming Subsystem for a Dynamically Scheduled Out-of-Order Core” March 2019. photo
- EDAA Outstanding Dissertation Award 2018 to my PhD student Anastasios Psarras in "Topic 2 – New directions in system-on-chip platforms co-design, novel emerging architectures and system-level management" for his PhD thesis "High-performance Networks-on-Chip". photoA photoB
- Second position to the timing optimization constest organized by the TAU workshop - ACM International Workshop on Timing Issues in the Specification and Synthesis of Digital Systems (TAU), March 2019. photo
- Hipeac technology transfer award "Network-on-Chip for Think Silicon’s ultra-low-power GPU", 2015.
- Best Paper Award - D track in Design Automation and Test in Europe conferene for the paper "PhaseNoC: TDM Scheduling at the Virtual-Channel Level for Efficient Network Traffic Isolation", March 2015. photoA photoB
Research Team
- Kosmas Alexandridis, PhD student
- Georgios Alexakis, PhD student
- Dimitrios Sekertzis, PhD student
PhD Theses Supervised
- Anastasios Psarras, "High-Performance Networks on Chip", 2017. pdf (EDAA Outstanding Dissertation Award)
- Ioannis Seitanidis , "Low-power Networks-on-Chip", 2018. pdf
- Dimitrios Mangiras, "Timing Optimization Techniques for the Scalable Physical Synthesis of Digital Integrated Circuits", 2022. pdf
- Dionysios Filippas, "Automated design of accelerators for machine learning applications", 2025. pdf
- Apostolos Stefanidis, "Reinforcement Learning for automated design of digital integrated circuits", 2025 pdf
- Christodoulos Peltekis, "Data Parallel Hardware Accelerators for Machine Learning Algorithms", 2025 pdf
- Vasileios Titopoulos, "Architecture of Energy-Efficient Vector Processors", Research completed, Phd thesis defence pending. 2026.
Teaching
- Integrated Circuits - Fall 2013 - 2024
- High-level synthesis for digital design - Fall 2020 - 2024
- VLSI systems - Spring 2013 - 2025
- Computer Organization - Fall 2021 - 2023
- Computer Architecture - Spring 2023 - 2024
- Systems-on-Chips - MSc/PhD course - Spring 2012-2025