|STORNAIUOLO LUCA||Cycle: XXXIII |
Section: Computer Science and Engineering
Tutor: DANIEL FLORIAN
Advisor: SCIUTO DONATELLA Major Research topic
:Co-designing Hardware/Software Infrastructures for Data Analytics and Artificial IntelligenceAbstract:
Starting from his M.Sc. thesis "Exploiting FPGA from Data Science Programming Languages", Luca Stornaiuolo is pursuing the work of creating optimized hardware and software infrastructure for data analytics. This involves the integration of hardware accelerators, such as FPGAs, within final users' applications for desktop or embedded systems. Raising the level of abstraction to create FPGA-based designs is an ongoing challenge to enhance the productivity for hardware developers and bridge the gap between hardware and software implementation approaches. Using hardware accelerators has shown to be a valuable solution also for Machine Learning algorithms, that are becoming pervasive in a large number of fields, from computer vision and robotics to finance and biotechnology. In this context, the main goal of Luca's research work is to develop an FPGA-based system optimized for data-intensive applications, like those based on Convolutional Neural Networks (CNNs), which has high performance and is easy to use by software developers and data scientists, who are used to working with high-level languages.