Beatrice Bussolino

Dottorando in Ingegneria Elettrica, Elettronica E Delle Comunicazioni , 35^o cycle (2019-2022)
Dipartimento di Elettronica e Telecomunicazioni (DET)

beatrice.bussolino@polito.it

Profilo

Dottorato di ricerca

Titolo della tesi

Techniques and Optimization Strategies for Efficient Hardware Acceleration of Neural Networks: Tap-Wisely-Quantized Winograd Algorithm and Capsule Networks

Argomento di ricerca

Hardware accelerators for Deep Neural Networks training and inference

Tutori

Maurizio Martina

Research presentation

Presentazione video

Interessi di ricerca

Big Data, Machine Learning, Neural Networks and Data Science

VLSI theory, design and applications

Biografia

Beatrice Bussolino received the B.Sc. and M.Sc. degrees in Electronic Engineering, in 2017 and 2019 respectively, at Politecnico di Torino, where she is now pursuing the Ph.D. degree in Electrical, Electronics and Communications Engineering under the supervision of Prof. Maurizio Martina. Her current research activity is in the field of Machine Learning and Deep Neural Networks (DNNs) in particular. Working in close collaboration with the Technical University of Vienna (TUWien) and Prof. Muhammad Shafique, the aim is the design of dedicated on-chip architectures for inference and training of DNNs, with a comprehensive approach from the model and its possible optimizations to the physical implementation.

Didattica

Insegnamenti

Corso di laurea magistrale

Integrated systems architecture. A.A. 2020/21, INGEGNERIA ELETTRONICA (ELECTRONIC ENGINEERING). Collaboratore del corso

MostraNascondi A.A. passati

Corso di laurea di 1° livello

Elettronica applicata. A.A. 2020/21, INGEGNERIA ELETTRONICA. Collaboratore del corso
Elettronica applicata. A.A. 2021/22, INGEGNERIA ELETTRONICA. Collaboratore del corso

MostraNascondi A.A. passati

Pubblicazioni

Pubblicazioni più recenti Vedi tutte le pubblicazioni su Porto@Iris

Marchisio, Alberto; Mrazek, Vojtech; Massa, Andrea; Bussolino, Beatrice; Martina, ... (2022)
RoHNAS: A Neural Architecture Search Framework with Conjoint Optimization for Adversarial Robustness and Hardware Efficiency of Convolutional and Capsule Networks. In: IEEE ACCESS, vol. 10, pp. 109043-109055. ISSN 2169-3536
Contributo su Rivista
Andri, Renzo; Bussolino, Beatrice; Cipolletta, Antonio; Cavigelli, Lukas; Wang, Zhe (2022)
Going Further With Winograd Convolutions: Tap-Wise Quantization for Efficient Inference on 4x4 Tile. In: IEEE/ACM International Symposium on Microarchitecture, Chicago (USA), 1-5 October 2022, pp. 582-598. ISBN: 978-1-6654-6272-3
Contributo in Atti di Convegno (Proceeding)
Aiello, Giuseppe; Bussolino, Beatrice; Valpreda, Emanuele; Ruo Roch, Massimo; Masera, ... (2022)
NLCMAP: A Framework for the Efficient Mapping of Non-Linear Convolutional Neural Networks on FPGA Accelerators. In: International Conference on Image Processing, Bordeaux, France, 16-19 October 2022. ISBN: 978-1-6654-9620-9
Contributo in Atti di Convegno (Proceeding)
Marchisio, Alberto; Bussolino, Beatrice; Salvati, Edoardo; Martina, Maurizio; Masera, ... (2022)
Enabling Capsule Networks at the Edge through Approximate Softmax and Squash Operations. In: ISLPED '22: ACM/IEEE International Symposium on Low Power Electronics and Design, Boston (USA), 1-2 August 2022. ISBN: 9781450393546
Contributo in Atti di Convegno (Proceeding)
Capra, M.; Bussolino, B.; Marchisio, A.; Masera, G.; Martina, M.; Shafique, M. (2020)
Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead. In: IEEE ACCESS, vol. 8, pp. 225134-225180. ISSN 2169-3536
Contributo su Rivista
Marchisio, Alberto; Massa, Andrea; Mrazek, Vojtech; Bussolino, Beatrice; Martina, ... (2020)
NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks. In: 2020 IEEE/ACM International Conference On Computer Aided Design (ICCAD), Virtual Event, USA, 2 - 5 Novembre, pp. 1-9. ISBN: 9781450380263
Contributo in Atti di Convegno (Proceeding)
Colucci, Alessio; Marchisio, Alberto; Bussolino, Beatrice; Mrazek, Voitech; Martina, ... (2020)
A Fast Design Space Exploration Framework for the Deep Learning Accelerators: Work-in-Progress. In: International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), Singapore, 20-25 September 2020, pp. 34-36. ISBN: 978-1-7281-9198-0
Contributo in Atti di Convegno (Proceeding)
Marchisio, Alberto; Bussolino, Beatrice; Colucci, Alessio; Martina, Maurizio; Masera, ... (2020)
Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks. In: 2020 57th ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, USA, USA, 20-24 July 2020, pp. 1-6. ISBN: 978-1-7281-1085-1
Contributo in Atti di Convegno (Proceeding)
Marchisio, Alberto; Bussolino, Beatrice; Colucci, Alessio; Hanif, Muhammad Abdullah; ... (2020)
FasTrCaps: An Integrated Framework for Fast yet Accurate Training of Capsule Networks. In: 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow (UK), 19-24 July 2020, pp. 1-8. ISBN: 978-1-7281-6926-2
Contributo in Atti di Convegno (Proceeding)
Capra, Maurizio; Bussolino, Beatrice; Marchisio, Alberto; Shafique, Muhammad; Masera, ... (2020)
An Updated Survey of Efficient Hardware Architectures for Accelerating Deep Convolutional Neural Networks. In: FUTURE INTERNET, vol. 12, pp. 113-134. ISSN 1999-5903
Contributo su Rivista

Mostra altre pubblicazioniMostra meno pubblicazioni