Beatrice Bussolino
Papers
1
Total Citations
5
H-Index
1
About
Beatrice Bussolino is a researcher at the forefront of hardware-software co-design for deep learning, with a particular focus on accelerating emerging neural network architectures. Her work addresses the critical gap between novel AI models and the hardware needed to run them efficiently. In her most cited work, "A Fast Design Space Exploration Framework for the Deep Learning Accelerators: Work-in-Progress" (2020), Bussolino tackles the computational challenges posed by Capsule Networks (CapsNets)—an advanced form of CNN that learns spatial relations and is invariant to transformations. She identifies that current accelerators are ill-suited for CapsNets' complex matrix operations, and proposes a framework to rapidly explore accelerator design spaces. This contribution is vital for enabling next-generation AI applications that require understanding of spatial hierarchies. With 5 citations on this paper alone, her work is gaining traction in the hardware-aware machine learning community. Bussolino’s research is instrumental in bridging the gap between algorithmic innovation and practical hardware implementation, making her a key figure in the evolution of efficient deep learning accelerators.
Research Focus
Key Achievements
Top Papers
- 1