T
Todor Stefanov
Researcher at Leiden University
Publications - 111
Citations - 2483
Todor Stefanov is an academic researcher from Leiden University. The author has contributed to research in topics: Scheduling (computing) & MPSoC. The author has an hindex of 24, co-authored 102 publications receiving 2312 citations. Previous affiliations of Todor Stefanov include Delft University of Technology.
Papers
More filters
Proceedings ArticleDOI
Optimization and deployment of CNNs at the edge: the ALOHA experience
Paolo Meloni,Daniela Loi,Paola Busia,Gianfranco Deriu,Andy D. Pimentel,Dolly Sapra,Todor Stefanov,Svetlana Minakova,Francesco Conti,Luca Benini,Maura Pintor,Battista Biggio,Bernhard Moser,Natalia Shepeleva,N. Fragoulis,Ilias Theodorakopoulos,Michael Masin,Francesca Palumbo +17 more
TL;DR: ALOHA is presented, an integrated tool flow that tries to facilitate the design of DL applications and their porting on embedded heterogenous architectures and aims at automating different design steps and reducing development costs.
Proceedings ArticleDOI
ALOHA: an architectural-aware framework for deep learning at the edge
Paolo Meloni,Daniela Loi,Gianfranco Deriu,Andy D. Pimentel,Dolly Sapra,Bernhard Moser,Natalia Shepeleva,Francesco Conti,Luca Benini,Oscar Ripolles,David Solans,Maura Pintor,Battista Biggio,Todor Stefanov,Svetlana Minakova,N. Fragoulis,Ilias Theodorakopoulos,Michael Masin,Francesca Palumbo +18 more
TL;DR: The ALOHA framework is described, that proposes a solution to the issue of effort and skills required to develop new DL models, or to adapt existing ones to new use-cases by means of an integrated tool flow that automates most phases of the development process.
Proceedings ArticleDOI
Efficient External Memory Interface for Multi-Processor Platforms Realized on FPGA Chips
TL;DR: A hierarchical memory system with a programmable controller to transfer data between external and on-chip memories using a DMA mechanism does not require arbitration which results in better overall performance.
Book ChapterDOI
System-Level Design Space Exploration of Dynamic Reconfigurable Architectures
TL;DR: A model is developed that can assist designers at the system-level DSE stage to explore the utilization of the reconfigurable resources and evaluate the relative impact of certain design choices and can be used to explore various design parameters by evaluating the system performance for different application-to-architecture mappings.
Y-Chart Based System Level Performance Analysis: An M-JPEG Case Study
TL;DR: This case study starts from a modified M-JPEG application and map this application onto a shared memory multi-processor architecture using Spade architecture and mapping languages.