Michele Favalli
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Michele Favalli has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
B. Ricco
38 shared publications- Piero Olivo
Dipartimento di Ingegneria, Universita degli Studi di Ferrara, Ferrara, Italy
30 shared publications C. Metra
DEIS - Univ. of Bologna
27 shared publications- Marcello Dalpasso
DEI - University of Padova
19 shared publications M. Damiani
10 shared publications- Davide Bertozzi
University of Manchester, Manchester, United Kingdom
5 shared publications
Latest Publications
- Open Access
- 4 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 31-38
Cross-Layer Analysis of Permanent Faults in Asynchronous NoCs for Neuromorphic Computing
- 4 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 31-38
Open AccessCross-Layer Analysis of Permanent Faults in Asynchronous NoCs for Neuromorphic Computing
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 121961-121977
Bias-Aware FP8 Format for Efficient Weight Representation in Post-Training Quantization
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 121961-121977
Open AccessBias-Aware FP8 Format for Efficient Weight Representation in Post-Training Quantization
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Golomb – Rice Coding for Efficient Compression of 8-Bit Floating- Point Data
Recent advancements in neural network (NN) hardware and software have introduced 8-bit Floating-Point (FP8) format as promising candidate for reducing memory usage and computational overhead. This article investigates the potential for further compressing FP8) data by applying Golomb-Rice (GR) coding. In the FP8 format, the 8 bits are allocated to represent the sign, exponent and mantissa using natural binary encoding. To achieve efficient compression, the GR code, a variable-length, lossless coding scheme, is applied exclusively to the exponent bits. In this way, data storage savings could be achieved, which can be very significant when the large amount of data has to be stored, such as storing NNs.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Golomb – Rice Coding for Efficient Compression of 8-Bit Floating- Point Data
Recent advancements in neural network (NN) hardware and software have introduced 8-bit Floating-Point (FP8) format as promising candidate for reducing memory usage and computational overhead. This article investigates the potential for further compressing FP8) data by applying Golomb-Rice (GR) coding. In the FP8 format, the 8 bits are allocated to represent the sign, exponent and mantissa using natural binary encoding. To achieve efficient compression, the GR code, a variable-length, lossless coding scheme, is applied exclusively to the exponent bits. In this way, data storage savings could be achieved, which can be very significant when the large amount of data has to be stored, such as storing NNs.
- Open Access
- 1 July 2025
- Published by Cambridge University Press (CUP) in Theory and Practice of Logic Programming
- Vol. 25 (4) , 522-539
Fine-Grained Timing Analysis of Digital Integrated Circuits in Answer Set Programming
In the design of integrated circuits, one critical metric is the maximum delay introduced by combinational modules within the circuit. This delay is crucial because it represents the time required to perform a computation: in an Arithmetic Logic Unit, it represents the maximum time taken by the circuit to perform an arithmetic operation. When such a circuit is part of a larger, synchronous system, like a CPU, the maximum delay directly impacts the maximum clock frequency of the entire system. Typically, hardware designers use static timing analysis to compute an upper bound of the maximum delay because it can be determined in polynomial time. However, relying on this upper bound can lead to suboptimal processor speeds, thereby missing performance opportunities. In this work, we tackle the challenging task of computing the actual maximum delay, rather than an approximate value. Since the problem is computationally hard, we model it in answer set programming (ASP), a logic language featuring extremely efficient solvers. We propose non-trivial encodings of the problem into ASP. Experimental results show that ASP is a viable solution to address complex problems in hardware design.
- 1 July 2025
- Published by Cambridge University Press (CUP) in Theory and Practice of Logic Programming
- Vol. 25 (4) , 522-539
Open AccessFine-Grained Timing Analysis of Digital Integrated Circuits in Answer Set Programming
In the design of integrated circuits, one critical metric is the maximum delay introduced by combinational modules within the circuit. This delay is crucial because it represents the time required to perform a computation: in an Arithmetic Logic Unit, it represents the maximum time taken by the circuit to perform an arithmetic operation. When such a circuit is part of a larger, synchronous system, like a CPU, the maximum delay directly impacts the maximum clock frequency of the entire system. Typically, hardware designers use static timing analysis to compute an upper bound of the maximum delay because it can be determined in polynomial time. However, relying on this upper bound can lead to suboptimal processor speeds, thereby missing performance opportunities. In this work, we tackle the challenging task of computing the actual maximum delay, rather than an approximate value. Since the problem is computationally hard, we model it in answer set programming (ASP), a logic language featuring extremely efficient solvers. We propose non-trivial encodings of the problem into ASP. Experimental results show that ASP is a viable solution to address complex problems in hardware design.
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 7700, 171-178
Efficient Resource-Aware Neural Architecture Search with a Neuro-Symbolic Approach
Hardware-aware Neural Architectural Search (NAS) is gaining momentum to enable the deployment of deep learning on edge devices with limited computing capabilities. Incorporating device-related objectives such as affordable floating point operations, latency, power, memory usage, etc. into the optimization process makes searching for the most efficient neural architecture more complicated, since both model accuracy and hardware cost should guide the search. The main concern with most state-of-the-art hardware-aware NAS strategies is that they propose for evaluation also trivially infeasible network models for the capabilities of the hardware platform at hand. Moreover, previously generated models are frequently not exploited to intelligently generate new ones, leading to prohibitive computational costs for practical relevance. This paper aims to boost the computational efficiency of hardware-aware NAS by means of a neuro-symbolic framework revolving around a Probabilistic Inductive Logic Programming module to define and exploit a set of symbolic rules. This component learns and refines the probabilities associated with the rules, allowing the framework to adapt and improve over time, thus quickly narrowing down the search space toward the most promising neural architectures.
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 7700, 171-178
Efficient Resource-Aware Neural Architecture Search with a Neuro-Symbolic Approach
Hardware-aware Neural Architectural Search (NAS) is gaining momentum to enable the deployment of deep learning on edge devices with limited computing capabilities. Incorporating device-related objectives such as affordable floating point operations, latency, power, memory usage, etc. into the optimization process makes searching for the most efficient neural architecture more complicated, since both model accuracy and hardware cost should guide the search. The main concern with most state-of-the-art hardware-aware NAS strategies is that they propose for evaluation also trivially infeasible network models for the capabilities of the hardware platform at hand. Moreover, previously generated models are frequently not exploited to intelligently generate new ones, leading to prohibitive computational costs for practical relevance. This paper aims to boost the computational efficiency of hardware-aware NAS by means of a neuro-symbolic framework revolving around a Probabilistic Inductive Logic Programming module to define and exploit a set of symbolic rules. This component learns and refines the probabilities associated with the rules, allowing the framework to adapt and improve over time, thus quickly narrowing down the search space toward the most promising neural architectures.
- 27 April 2022
- Published by Springer Nature
- p. 505-522
The Challenge of Classification Confidence Estimation in Dynamically-Adaptive Neural Networks
- 27 April 2022
- Published by Springer Nature
- p. 505-522
The Challenge of Classification Confidence Estimation in Dynamically-Adaptive Neural Networks
- Open Access
- 27 January 2022
- Published by MDPI AG in Electronics
- Vol. 11 (3) , 382
Techniques for SAT-Based Boolean Reasoning on Multiple Faults Affecting Logic Cells and Interconnects in Digital ICs
We propose a new method for SAT-based Boolean reasoning on multiple defects in digital ICs. Although it does not explicitly consider a specific fault model such as model-based techniques, it allows us to consider more realistic cases than model-free approaches. In particular, it can be used to account for (a) faults resulting in monotonic errors at the output of a cell and (b) faults, such as breaks or bridgings, that may corrupt the propagation of a signal from its fan-out branches. The model can be used for either standard gates or more complex combinational modules. Examples are shown for applications requiring the consideration of multiple defects such as fault diagnosis and reliability analysis. The feasibility of the proposed approach is assessed by results on a set of combinational benchmarks.
- 27 January 2022
- Published by MDPI AG in Electronics
- Vol. 11 (3) , 382
Open AccessTechniques for SAT-Based Boolean Reasoning on Multiple Faults Affecting Logic Cells and Interconnects in Digital ICs
We propose a new method for SAT-based Boolean reasoning on multiple defects in digital ICs. Although it does not explicitly consider a specific fault model such as model-based techniques, it allows us to consider more realistic cases than model-free approaches. In particular, it can be used to account for (a) faults resulting in monotonic errors at the output of a cell and (b) faults, such as breaks or bridgings, that may corrupt the propagation of a signal from its fan-out branches. The model can be used for either standard gates or more complex combinational modules. Examples are shown for applications requiring the consideration of multiple defects such as fault diagnosis and reliability analysis. The feasibility of the proposed approach is assessed by results on a set of combinational benchmarks.
- Open Access
- 27 June 2021
- Published by MDPI AG in Micromachines
- Vol. 12 (7) , 759
A Scalable Bidimensional Randomization Scheme for TLC 3D NAND Flash Memories
Data randomization has been a widely adopted Flash Signal Processing technique for reducing or suppressing errors since the inception of mass storage platforms based on planar NAND Flash technology. However, the paradigm change represented by the 3D memory integration concept has complicated the randomization task due to the increased dimensions of the memory array, especially along the bitlines. In this work, we propose an easy to implement, cost effective, and fully scalable with memory dimensions, randomization scheme that guarantees optimal randomization along the wordline and the bitline dimensions. At the same time, we guarantee an upper bound on the maximum length of consecutive ones and zeros along the bitline to improve the memory reliability. Our method has been validated on commercial off-the-shelf TLC 3D NAND Flash memory with respect to the Raw Bit Error Rate metric extracted in different memory working conditions.
- 27 June 2021
- Published by MDPI AG in Micromachines
- Vol. 12 (7) , 759
Open AccessA Scalable Bidimensional Randomization Scheme for TLC 3D NAND Flash Memories
Data randomization has been a widely adopted Flash Signal Processing technique for reducing or suppressing errors since the inception of mass storage platforms based on planar NAND Flash technology. However, the paradigm change represented by the 3D memory integration concept has complicated the randomization task due to the increased dimensions of the memory array, especially along the bitlines. In this work, we propose an easy to implement, cost effective, and fully scalable with memory dimensions, randomization scheme that guarantees optimal randomization along the wordline and the bitline dimensions. At the same time, we guarantee an upper bound on the maximum length of consecutive ones and zeros along the bitline to improve the memory reliability. Our method has been validated on commercial off-the-shelf TLC 3D NAND Flash memory with respect to the Raw Bit Error Rate metric extracted in different memory working conditions.
- 3 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
A built-in self-testing framework for asynchronous bundled-data NoC switches resilient to delay variations
Most multi- and many-core integrated systems are currently designed by following a globally asynchronous locally synchronous paradigm. Asynchronous interconnection networks are promising candidates to interconnect IP cores operating at potentially different frequencies. Nevertheless, post-fabrication testing is a big challenge to bring asynchronous NoCs to the market due to a lack of testing methodologies and support for them. In particular, the unpredictable delay variability introduced by the manufacturing process may differentiate the delay of nominally-balanced I/O timing paths, thus making the order of the input patterns unpredictable and precluding the correct behaviour of signature-based test compactors. This paper tackles this challenge by proposing a testing framework for asynchronous NoCs which works effectively despite delay variations in and across timing paths of the NoC under test. Moreover, in order to mitigate the growing test application costs in modern ICs, we come up with a built-in self-testing infrastructure which automatically controls and delivers the outcome of the testing process without the intervention of an external automatic test equipment (ATE).
- 3 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
A built-in self-testing framework for asynchronous bundled-data NoC switches resilient to delay variations
Most multi- and many-core integrated systems are currently designed by following a globally asynchronous locally synchronous paradigm. Asynchronous interconnection networks are promising candidates to interconnect IP cores operating at potentially different frequencies. Nevertheless, post-fabrication testing is a big challenge to bring asynchronous NoCs to the market due to a lack of testing methodologies and support for them. In particular, the unpredictable delay variability introduced by the manufacturing process may differentiate the delay of nominally-balanced I/O timing paths, thus making the order of the input patterns unpredictable and precluding the correct behaviour of signature-based test compactors. This paper tackles this challenge by proposing a testing framework for asynchronous NoCs which works effectively despite delay variations in and across timing paths of the NoC under test. Moreover, in order to mitigate the growing test application costs in modern ICs, we come up with a built-in self-testing infrastructure which automatically controls and delivers the outcome of the testing process without the intervention of an external automatic test equipment (ATE).