Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Alberto Marchisio

Alberto Marchisio

Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead


Dec 21, 2020
Maurizio Capra, Beatrice Bussolino, Alberto Marchisio, Guido Masera, Maurizio Martina, Muhammad Shafique

Figure 1 for Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
Figure 2 for Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
Figure 3 for Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
Figure 4 for Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
* Accepted for publication in IEEE Access 
  Access Paper or Ask Questions

MLComp: A Methodology for Machine Learning-based Performance Estimation and Adaptive Selection of Pareto-Optimal Compiler Optimization Sequences


Dec 11, 2020
Alessio Colucci, Dávid Juhász, Martin Mosbeck, Alberto Marchisio, Semeen Rehman, Manfred Kreutzer, Guenther Nadbath, Axel Jantsch, Muhammad Shafique

Figure 1 for MLComp: A Methodology for Machine Learning-based Performance Estimation and Adaptive Selection of Pareto-Optimal Compiler Optimization Sequences
Figure 2 for MLComp: A Methodology for Machine Learning-based Performance Estimation and Adaptive Selection of Pareto-Optimal Compiler Optimization Sequences
Figure 3 for MLComp: A Methodology for Machine Learning-based Performance Estimation and Adaptive Selection of Pareto-Optimal Compiler Optimization Sequences
Figure 4 for MLComp: A Methodology for Machine Learning-based Performance Estimation and Adaptive Selection of Pareto-Optimal Compiler Optimization Sequences
* Accepted for publication at the 24th IEEE/ACM Design, Automation and Test in Europe (DATE'21) Conference, February, 2021 
  Access Paper or Ask Questions

Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters


Dec 09, 2020
Rida El-Allami, Alberto Marchisio, Muhammad Shafique, Ihsen Alouani

Figure 1 for Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters
Figure 2 for Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters
Figure 3 for Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters
Figure 4 for Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters
* Accepted for publication at the 24th Design, Automation and Test in Europe (DATE'21), February, 2021 
  Access Paper or Ask Questions

DESCNet: Developing Efficient Scratchpad Memories for Capsule Network Hardware


Oct 12, 2020
Alberto Marchisio, Vojtech Mrazek, Muhammad Abdullah Hanif, Muhammad Shafique

Figure 1 for DESCNet: Developing Efficient Scratchpad Memories for Capsule Network Hardware
Figure 2 for DESCNet: Developing Efficient Scratchpad Memories for Capsule Network Hardware
Figure 3 for DESCNet: Developing Efficient Scratchpad Memories for Capsule Network Hardware
Figure 4 for DESCNet: Developing Efficient Scratchpad Memories for Capsule Network Hardware
* Accepted for publication at the IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 
  Access Paper or Ask Questions

NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks


Aug 19, 2020
Alberto Marchisio, Andrea Massa, Vojtech Mrazek, Beatrice Bussolino, Maurizio Martina, Muhammad Shafique

Figure 1 for NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks
Figure 2 for NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks
Figure 3 for NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks
Figure 4 for NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks
* To appear at the IEEE/ACM International Conference on Computer-Aided Design (ICCAD '20), November 2-5, 2020, Virtual Event, USA 
  Access Paper or Ask Questions

NeuroAttack: Undermining Spiking Neural Networks Security through Externally Triggered Bit-Flips


May 16, 2020
Valerio Venceslai, Alberto Marchisio, Ihsen Alouani, Maurizio Martina, Muhammad Shafique

Figure 1 for NeuroAttack: Undermining Spiking Neural Networks Security through Externally Triggered Bit-Flips
Figure 2 for NeuroAttack: Undermining Spiking Neural Networks Security through Externally Triggered Bit-Flips
Figure 3 for NeuroAttack: Undermining Spiking Neural Networks Security through Externally Triggered Bit-Flips
Figure 4 for NeuroAttack: Undermining Spiking Neural Networks Security through Externally Triggered Bit-Flips
* Accepted for publication at the 2020 International Joint Conference on Neural Networks (IJCNN) 
  Access Paper or Ask Questions

Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks


Apr 17, 2020
Alberto Marchisio, Beatrice Bussolino, Alessio Colucci, Maurizio Martina, Guido Masera, Muhammad Shafique

Figure 1 for Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks
Figure 2 for Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks
Figure 3 for Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks
Figure 4 for Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks
* Accepted for publication at Design Automation Conference 2020 (DAC 2020) 
  Access Paper or Ask Questions

ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations


Dec 02, 2019
Alberto Marchisio, Vojtech Mrazek, Muhammad Abudllah Hanif, Muhammad Shafique

Figure 1 for ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations
Figure 2 for ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations
Figure 3 for ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations
Figure 4 for ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations
* To appear at the 23rd Design, Automation and Test in Europe (DATE 2020). Grenoble, France 
  Access Paper or Ask Questions

X-TrainCaps: Accelerated Training of Capsule Nets through Lightweight Software Optimizations


May 24, 2019
Alberto Marchisio, Beatrice Bussolino, Alessio Colucci, Muhammad Abdullah Hanif, Maurizio Martina, Guido Masera, Muhammad Shafique

Figure 1 for X-TrainCaps: Accelerated Training of Capsule Nets through Lightweight Software Optimizations
Figure 2 for X-TrainCaps: Accelerated Training of Capsule Nets through Lightweight Software Optimizations
Figure 3 for X-TrainCaps: Accelerated Training of Capsule Nets through Lightweight Software Optimizations
Figure 4 for X-TrainCaps: Accelerated Training of Capsule Nets through Lightweight Software Optimizations

  Access Paper or Ask Questions

CapStore: Energy-Efficient Design and Management of the On-Chip Memory for CapsuleNet Inference Accelerators


Apr 12, 2019
Alberto Marchisio, Muhammad Abdullah Hanif, Mohammad Taghi Teimoori, Muhammad Shafique

Figure 1 for CapStore: Energy-Efficient Design and Management of the On-Chip Memory for CapsuleNet Inference Accelerators
Figure 2 for CapStore: Energy-Efficient Design and Management of the On-Chip Memory for CapsuleNet Inference Accelerators
Figure 3 for CapStore: Energy-Efficient Design and Management of the On-Chip Memory for CapsuleNet Inference Accelerators
Figure 4 for CapStore: Energy-Efficient Design and Management of the On-Chip Memory for CapsuleNet Inference Accelerators

  Access Paper or Ask Questions

SNN under Attack: are Spiking Deep Belief Networks vulnerable to Adversarial Examples?


Feb 04, 2019
Alberto Marchisio, Giorgio Nanfa, Faiq Khalid, Muhammad Abdullah Hanif, Maurizio Martina, Muhammad Shafique

Figure 1 for SNN under Attack: are Spiking Deep Belief Networks vulnerable to Adversarial Examples?
Figure 2 for SNN under Attack: are Spiking Deep Belief Networks vulnerable to Adversarial Examples?
Figure 3 for SNN under Attack: are Spiking Deep Belief Networks vulnerable to Adversarial Examples?
Figure 4 for SNN under Attack: are Spiking Deep Belief Networks vulnerable to Adversarial Examples?

  Access Paper or Ask Questions

CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks


Jan 28, 2019
Alberto Marchisio, Giorgio Nanfa, Faiq Khalid, Muhammad Abdullah Hanif, Maurizio Martina, Muhammad Shafique

Figure 1 for CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks
Figure 2 for CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks
Figure 3 for CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks
Figure 4 for CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks

  Access Paper or Ask Questions