Typology
operator:
and / or
Language
operator:
and / or
Date
operator:
and / or
Rights
operator:
and / or
2022
Conference article
Open Access
Foreseeing the impact of the proposed AI Act on the sustainability and safety of critical infrastructures
Sovrano F., Masetti G.
The AI Act has been recently proposed by the European Commission to regulate the use of AI in the EU, especially on high-risk applications, i.e. systems intended to be used as safety components in the management and operation of road traffic and the supply of water, gas, heating and electricity. On the other hand, IEC 61508, one of the most adopted international standards for safety-critical electronic components, seem to mostly forbid the use of AI in such systems. Given this conflict between IEC 61508 and the proposed AI Act, also stressed by the fact that IEC 61508 is not an harmonised European standard, with the present paper we study and analyse what is going to happen to industry after the entry into force of the AI Act. More in detail we focus on how the proposed AI Act might positively impact on the sustainability of critical infrastructures by allowing the use of AI on an industry where it was previously forbidden. To do so, we provide several examples of AI-based solutions falling under the umbrella of IEC 61508 that might have a positive impact on sustainability in alignment with the current long-term goals of the EU and the Sustainable Development Goals of the United Nations, i.e. 1) affordable and clean energy, 2) sustainable cities and communities.Source: ICEGOV 2022 - 15th International Conference on Theory and Practice of Electronic Governance, pp. 492–498, Guimarães, Portugal, 04-07/10/2022
DOI: 10.1145/3560107.3560253
Metrics:
Sovrano F., Masetti G.
The AI Act has been recently proposed by the European Commission to regulate the use of AI in the EU, especially on high-risk applications, i.e. systems intended to be used as safety components in the management and operation of road traffic and the supply of water, gas, heating and electricity. On the other hand, IEC 61508, one of the most adopted international standards for safety-critical electronic components, seem to mostly forbid the use of AI in such systems. Given this conflict between IEC 61508 and the proposed AI Act, also stressed by the fact that IEC 61508 is not an harmonised European standard, with the present paper we study and analyse what is going to happen to industry after the entry into force of the AI Act. More in detail we focus on how the proposed AI Act might positively impact on the sustainability of critical infrastructures by allowing the use of AI on an industry where it was previously forbidden. To do so, we provide several examples of AI-based solutions falling under the umbrella of IEC 61508 that might have a positive impact on sustainability in alignment with the current long-term goals of the EU and the Sustainable Development Goals of the United Nations, i.e. 1) affordable and clean energy, 2) sustainable cities and communities.Source: ICEGOV 2022 - 15th International Conference on Theory and Practice of Electronic Governance, pp. 492–498, Guimarães, Portugal, 04-07/10/2022
DOI: 10.1145/3560107.3560253
Metrics:
See at:
ISTI Repository 
| dl.acm.org 
| CNR ExploRA
2022
Journal article
Open Access
Groundwater modeling with process-based and data-driven approaches in the context of climate change
Menichini M., Franceschi L., Raco B., Masetti G., Scozzari A., Doveri M.
In the context of climate change, the correct management of groundwater, which is strategic for meeting water needs, becomes essential. Groundwater modeling is particularly crucial for the sustainable and efficient management of groundwater. This manuscript provides different types of modeling according to data availability and features of three porous aquifer systems in Italy (Empoli, Magra, and Brenta systems). The models calibrated on robust time series enabled the performing of forecast simulations capable of representing the quantitative and qualitative response to expected climate regimes. For the Empoli aquifer, the process-based models highlighted the system's ability to mitigate the effects of dry climate conditions thanks to its storage capability. The data-driven models concerning the Brenta foothill aquifer pointed out the high sensitivity of the system to climate extremes, thus suggesting the need for specific water management actions. The integrated datadriven/process-based approach developed for the Magra Valley aquifer remarked that the water quantity and quality effects are tied to certain boundary conditions over dry climate periods. This work shows that, for groundwater modeling, the choice of the suitable approach is mandatory, and it mainly depends on the specific aquifer features that result in different ways to be sensitive to climate. This manuscript also provides a novel outcome involving the integrated approach wherein it is a very efficient tool for forecasting modeling when boundary conditions, which significantly affect the behavior of such systems, are subjected to evolve under expected climate scenarios.Source: Water (Basel) 14 (2022). doi:10.3390/w14233956
DOI: 10.3390/w14233956
Metrics:
Menichini M., Franceschi L., Raco B., Masetti G., Scozzari A., Doveri M.
In the context of climate change, the correct management of groundwater, which is strategic for meeting water needs, becomes essential. Groundwater modeling is particularly crucial for the sustainable and efficient management of groundwater. This manuscript provides different types of modeling according to data availability and features of three porous aquifer systems in Italy (Empoli, Magra, and Brenta systems). The models calibrated on robust time series enabled the performing of forecast simulations capable of representing the quantitative and qualitative response to expected climate regimes. For the Empoli aquifer, the process-based models highlighted the system's ability to mitigate the effects of dry climate conditions thanks to its storage capability. The data-driven models concerning the Brenta foothill aquifer pointed out the high sensitivity of the system to climate extremes, thus suggesting the need for specific water management actions. The integrated datadriven/process-based approach developed for the Magra Valley aquifer remarked that the water quantity and quality effects are tied to certain boundary conditions over dry climate periods. This work shows that, for groundwater modeling, the choice of the suitable approach is mandatory, and it mainly depends on the specific aquifer features that result in different ways to be sensitive to climate. This manuscript also provides a novel outcome involving the integrated approach wherein it is a very efficient tool for forecasting modeling when boundary conditions, which significantly affect the behavior of such systems, are subjected to evolve under expected climate scenarios.Source: Water (Basel) 14 (2022). doi:10.3390/w14233956
DOI: 10.3390/w14233956
Metrics:
See at:
ISTI Repository | www.mdpi.com | CNR ExploRA
2022
Conference article
Open Access
Solution bundles of Markov performability models through adaptive cross approximation
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
A technique to approximate solution bundles, i.e., solutions of a parametric model where parameters are treated as independent variables instead of constants, is presented for Markov models. Analyses based on an approximated solution bundle are more efficient than those that solve the model for all combinations of parameters' values separately. In this paper the idea is to properly adapt low rank tensor approximation techniques, and in particular Adaptive Cross Approximation, to the evaluation of performability attributes. Application on exemplary case studies confirms the advantages of the new solution technique with respect to solving the model for all time and parameters' combinations.Source: DSN 2022 - 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks, pp. 381–392, Baltimora, USA, 27-30/06/2022
DOI: 10.1109/dsn53405.2022.00046
Metrics:
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
A technique to approximate solution bundles, i.e., solutions of a parametric model where parameters are treated as independent variables instead of constants, is presented for Markov models. Analyses based on an approximated solution bundle are more efficient than those that solve the model for all combinations of parameters' values separately. In this paper the idea is to properly adapt low rank tensor approximation techniques, and in particular Adaptive Cross Approximation, to the evaluation of performability attributes. Application on exemplary case studies confirms the advantages of the new solution technique with respect to solving the model for all time and parameters' combinations.Source: DSN 2022 - 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks, pp. 381–392, Baltimora, USA, 27-30/06/2022
DOI: 10.1109/dsn53405.2022.00046
Metrics:
See at:
ISTI Repository | doi.org | ieeexplore.ieee.org | CNR ExploRA
2022
Journal article
Open Access
Implicit reward structures for implicit reliability models
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
A new methodology for effective definition and efficient evaluation of dependability-related properties is proposed. The analysis targets the systems composed of a large number of components, each one modeled implicitly through high-level formalisms, such as stochastic Petri nets. Since the component models are implicit, the reward structure that characterizes the dependability properties has to be implicit as well. Therefore, we present a new formalism to specify those reward structures. The focus here is on component models that can be mapped to stochastic automata with one or several absorbing states so that the system model can be mapped to a stochastic automata network with one or several absorbing states. Correspondingly, the new reward structure defined on each component's model is mapped to a reward vector so that the dependability-related properties of the system are expressed through a newly introduced measure defined starting from those reward vectors. A simple, yet representative, case study is adopted to show the feasibility of the method.Source: IEEE transactions on reliability (2022). doi:10.1109/TR.2022.3190915
DOI: 10.1109/tr.2022.3190915
Metrics:
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
A new methodology for effective definition and efficient evaluation of dependability-related properties is proposed. The analysis targets the systems composed of a large number of components, each one modeled implicitly through high-level formalisms, such as stochastic Petri nets. Since the component models are implicit, the reward structure that characterizes the dependability properties has to be implicit as well. Therefore, we present a new formalism to specify those reward structures. The focus here is on component models that can be mapped to stochastic automata with one or several absorbing states so that the system model can be mapped to a stochastic automata network with one or several absorbing states. Correspondingly, the new reward structure defined on each component's model is mapped to a reward vector so that the dependability-related properties of the system are expressed through a newly introduced measure defined starting from those reward vectors. A simple, yet representative, case study is adopted to show the feasibility of the method.Source: IEEE transactions on reliability (2022). doi:10.1109/TR.2022.3190915
DOI: 10.1109/tr.2022.3190915
Metrics:
See at:
ISTI Repository | ieeexplore.ieee.org | CNR ExploRA
2022
Journal article
Open Access
TAPAS: a tool for stochastic evaluation of large interdependent composed models with absorbing states
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
TAPAS is a new tool for efficient evaluation of dependability and performability attributes of systems composed of many interconnected components. The tool solves homogeneous continuous time Markov chains described by stochastic automata network models structured in submodels with absorbing states. The measures of interest are defined by a reward structure based on submodels composed through transition-based synchronization. The tool has been conceived in a modular and flexible fashion, to easily accommodate new features. Currently, it implements an array of state-based solvers that addresses the state explosion problem through powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. A simple, yet representative, case study is adopted, to present the tool and to show the feasibility of the supported methods, in particular frommemory consumption point of view.Source: Performance evaluation review 49 (2022): 41–46. doi:10.1145/3543146.3543157
DOI: 10.1145/3543146.3543157
Metrics:
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
TAPAS is a new tool for efficient evaluation of dependability and performability attributes of systems composed of many interconnected components. The tool solves homogeneous continuous time Markov chains described by stochastic automata network models structured in submodels with absorbing states. The measures of interest are defined by a reward structure based on submodels composed through transition-based synchronization. The tool has been conceived in a modular and flexible fashion, to easily accommodate new features. Currently, it implements an array of state-based solvers that addresses the state explosion problem through powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. A simple, yet representative, case study is adopted, to present the tool and to show the feasibility of the supported methods, in particular frommemory consumption point of view.Source: Performance evaluation review 49 (2022): 41–46. doi:10.1145/3543146.3543157
DOI: 10.1145/3543146.3543157
Metrics:
See at:
ISTI Repository | dl.acm.org | ACM SIGMETRICS Performance Evaluation Review | CNR ExploRA
2022
Journal article
Open Access
Redundancy-based intrusion tolerance approaches moving from classical fault tolerance methods
Di Giandomenico F., Masetti G., Chiaradonna S.
Borrowing from well known fault tolerant approaches based on redundancy to mask the effect of faults, redundancy-based intrusion tolerance schemes are proposed in this paper, where redundancy of ICT components is exploited as a first defense line against a subset of compromised components within the redundant set, due to cyberattacks. Features to enhance defense and tolerance capabilities are first discussed, covering diversity-based redundancy, confusion techniques, protection mechanisms, locality policies and rejuvenation phases. Then, a set of intrusion tolerance variations of classical fault tolerant schemes (including N Version Programming and Recovery Block, as well as a few hybrid approaches) is proposed, by enriching each original scheme with one or more of the previously introduced defense mechanisms. As a practical support to the system designer in making an appropriate choice among the available solutions, for each developed scheme a schematic summary is provided, in terms of resources and defense facilities needed to tolerate f value failures and k omission failures, as well as observations regarding time requirements. To provide an example of more detailed analysis, useful to set up an appropriate intrusion tolerance configuration, a trade-off study between cost and additional redundancy employed for confusion purposes is also carried out.Source: International Journal of Applied Mathematics and Computer Science 32 (2022): 701–719. doi:10.34768/amcs-2022-0048
DOI: 10.34768/amcs-2022-0048
Project(s): BIECO
Metrics:
Di Giandomenico F., Masetti G., Chiaradonna S.
Borrowing from well known fault tolerant approaches based on redundancy to mask the effect of faults, redundancy-based intrusion tolerance schemes are proposed in this paper, where redundancy of ICT components is exploited as a first defense line against a subset of compromised components within the redundant set, due to cyberattacks. Features to enhance defense and tolerance capabilities are first discussed, covering diversity-based redundancy, confusion techniques, protection mechanisms, locality policies and rejuvenation phases. Then, a set of intrusion tolerance variations of classical fault tolerant schemes (including N Version Programming and Recovery Block, as well as a few hybrid approaches) is proposed, by enriching each original scheme with one or more of the previously introduced defense mechanisms. As a practical support to the system designer in making an appropriate choice among the available solutions, for each developed scheme a schematic summary is provided, in terms of resources and defense facilities needed to tolerate f value failures and k omission failures, as well as observations regarding time requirements. To provide an example of more detailed analysis, useful to set up an appropriate intrusion tolerance configuration, a trade-off study between cost and additional redundancy employed for confusion purposes is also carried out.Source: International Journal of Applied Mathematics and Computer Science 32 (2022): 701–719. doi:10.34768/amcs-2022-0048
DOI: 10.34768/amcs-2022-0048
Project(s): BIECO
Metrics:
See at:
ISTI Repository | www.amcs.uz.zgora.pl | doi.org | CNR ExploRA
2022
Conference article
Open Access
Random bad state estimator to address false data injection in critical infrastructures
Masetti G., Chiaradonna S., Robol L., Di Giandomenico F.
Given their crucial role for a society and economy, an essential component of critical infrastructures is the Bad State Estimator (BSE), responsible for detecting malfunctions affecting elements of the physical infrastructure. In the past, the BSE has been conceived to mainly cope with accidental faults, under assumptions characterizing their occurrence. However, evolution of the addressed systems category consisting in pervasiveness of ICT-based control towards increasing smartness, paired with the openness of the operational environment, contributed to expose critical infrastructures to intentional attacks, e.g. exploited through False Data Injection (FDI). In the flow of studies focusing on enhancements of the traditional BSE to account for FDI attacks, this paper proposes a new solution that introduces randomness elements in the diagnosis process, to improve detection abilities and mitigate potentially catastrophic common-mode errors. Differently from existing alternatives, the strength of this new technique is that it does not require any additional components or alternative source of information with respect to the classic BSE. Numerical experiments conducted on two IEEE transmission grid tests, taken as representative use cases, show the applicability and benefits of the new solution.Source: PRDC 2022 - 27th IEEE Pacific Rim International Symposium on Dependable Computing, pp. 98–108, Beijing, China, 28/11/2022 - 01/12/2022
DOI: 10.1109/prdc55274.2022.00024
Metrics:
Masetti G., Chiaradonna S., Robol L., Di Giandomenico F.
Given their crucial role for a society and economy, an essential component of critical infrastructures is the Bad State Estimator (BSE), responsible for detecting malfunctions affecting elements of the physical infrastructure. In the past, the BSE has been conceived to mainly cope with accidental faults, under assumptions characterizing their occurrence. However, evolution of the addressed systems category consisting in pervasiveness of ICT-based control towards increasing smartness, paired with the openness of the operational environment, contributed to expose critical infrastructures to intentional attacks, e.g. exploited through False Data Injection (FDI). In the flow of studies focusing on enhancements of the traditional BSE to account for FDI attacks, this paper proposes a new solution that introduces randomness elements in the diagnosis process, to improve detection abilities and mitigate potentially catastrophic common-mode errors. Differently from existing alternatives, the strength of this new technique is that it does not require any additional components or alternative source of information with respect to the classic BSE. Numerical experiments conducted on two IEEE transmission grid tests, taken as representative use cases, show the applicability and benefits of the new solution.Source: PRDC 2022 - 27th IEEE Pacific Rim International Symposium on Dependable Computing, pp. 98–108, Beijing, China, 28/11/2022 - 01/12/2022
DOI: 10.1109/prdc55274.2022.00024
Metrics:
See at:
ISTI Repository | doi.org | ieeexplore.ieee.org | CNR ExploRA
2021
Journal article
Open Access
Enhancing sustainability of the railway infrastructure: Trading energy saving and unavailability through efficient switch heating policies
Chiaradonna S., Masetti G., Di Giandomenico F., Righetti F., Vallati C.
Railway is currently envisioned as the most promising transportation system for both people and freight to reduce atmospheric emission and combat climate change. In this context, ensuring the energy efficiency of the railway systems is paramount in order to sustain their future expandability with minimum carbon footprint. Recent advancements in computing and communication technologies are expected to play a significant role to enable novel integrated control and management strategies in which heterogeneous data is exploited to noticeably increase energy efficiency. In this paper we focus on exploiting the convergence of heterogeneous information to improve energy efficiency of railway systems, in particular on the heating system for the railroad switches, one of the major energy intensive components. To this aim, we define new policies to efficiently manage the heating of these switches exploiting also external information such as weather and forecast data. In order to assess the performance of each strategy, a stochastic model representing the structure and operation of the railroad switch heating system and environmental conditions (both weather profiles and specific failure events) has been developed and exercised in a variety of representative scenarios. The obtained results allow to understand both strengths and limitations of each energy management policy, and serves as a useful support to make the choice of the best technique to employ to save on energy consumption, given the system conditions at hand.Source: Sustainable computing: informatics and systems (Print) 30 (2021). doi:10.1016/j.suscom.2021.100519
DOI: 10.1016/j.suscom.2021.100519
Metrics:
Chiaradonna S., Masetti G., Di Giandomenico F., Righetti F., Vallati C.
Railway is currently envisioned as the most promising transportation system for both people and freight to reduce atmospheric emission and combat climate change. In this context, ensuring the energy efficiency of the railway systems is paramount in order to sustain their future expandability with minimum carbon footprint. Recent advancements in computing and communication technologies are expected to play a significant role to enable novel integrated control and management strategies in which heterogeneous data is exploited to noticeably increase energy efficiency. In this paper we focus on exploiting the convergence of heterogeneous information to improve energy efficiency of railway systems, in particular on the heating system for the railroad switches, one of the major energy intensive components. To this aim, we define new policies to efficiently manage the heating of these switches exploiting also external information such as weather and forecast data. In order to assess the performance of each strategy, a stochastic model representing the structure and operation of the railroad switch heating system and environmental conditions (both weather profiles and specific failure events) has been developed and exercised in a variety of representative scenarios. The obtained results allow to understand both strengths and limitations of each energy management policy, and serves as a useful support to make the choice of the best technique to employ to save on energy consumption, given the system conditions at hand.Source: Sustainable computing: informatics and systems (Print) 30 (2021). doi:10.1016/j.suscom.2021.100519
DOI: 10.1016/j.suscom.2021.100519
Metrics:
See at:
ISTI Repository | Sustainable Computing Informatics and Systems | www.sciencedirect.com | CNR ExploRA
2021
Journal article
Open Access
On identity-aware replication in stochastic modeling for simulation-based dependability analysis of large interconnected systems
Chiaradonna S., Di Giandomenico F., Masetti G.
This paper focuses on the generation of stochastic models for dependability and performability analysis, through mechanisms for the automatic replication of template models when identity of replicas cannot be anonymous. The major objective of this work is to support the modeler in selecting the most appropriate replication mechanism, given the characteristics of the system under analysis. To this purpose, three most used solutions to identity-aware replication are considered and a formal framework to allow representing them in a consistent way is first defined. Then, a comparison of their behavior is extensively carried out, with focus on efficiency, both from a theoretical perspective and from a quantitative viewpoint. For the latter, a specific implementation of the considered replication mechanisms in the Möbius modeling environment and a case study representative of realistic interconnected infrastructures are developed.Source: Performance evaluation 147 (2021). doi:10.1016/j.peva.2021.102192
DOI: 10.1016/j.peva.2021.102192
Metrics:
Chiaradonna S., Di Giandomenico F., Masetti G.
This paper focuses on the generation of stochastic models for dependability and performability analysis, through mechanisms for the automatic replication of template models when identity of replicas cannot be anonymous. The major objective of this work is to support the modeler in selecting the most appropriate replication mechanism, given the characteristics of the system under analysis. To this purpose, three most used solutions to identity-aware replication are considered and a formal framework to allow representing them in a consistent way is first defined. Then, a comparison of their behavior is extensively carried out, with focus on efficiency, both from a theoretical perspective and from a quantitative viewpoint. For the latter, a specific implementation of the considered replication mechanisms in the Möbius modeling environment and a case study representative of realistic interconnected infrastructures are developed.Source: Performance evaluation 147 (2021). doi:10.1016/j.peva.2021.102192
DOI: 10.1016/j.peva.2021.102192
Metrics:
See at:
ISTI Repository | Performance Evaluation | www.sciencedirect.com | CNR ExploRA
2021
Contribution to conference
Open Access
TAPAS: a tool for stochastic evaluation of large interdependent composed models with absorbing states
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
TAPAS is a new tool for efficient evaluation of dependability and performability attributes of systems composed of many interconnected components. The tool solves homogeneous continuous time Markov chains described by stochastic automata network models structured in submodels with absorbing states. The measures of interest are defined by a reward structure based on submodels composed through transition-based synchronization. The tool has been conceived in a modular and flexible fashion, to easily accommodate new features. Currently, it implements an array of state-based solvers that addresses the state explosion problem through powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. A simple, yet representative, case study is adopted, to present the tool and to show the feasibility of the supported methods, in particular from memory consumption point of view.Source: TOSME Workshop - Tools for Stochastic Modelling and Evaluation (performance, dependability, security and verification), Online workshop, 12/11/2021
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
TAPAS is a new tool for efficient evaluation of dependability and performability attributes of systems composed of many interconnected components. The tool solves homogeneous continuous time Markov chains described by stochastic automata network models structured in submodels with absorbing states. The measures of interest are defined by a reward structure based on submodels composed through transition-based synchronization. The tool has been conceived in a modular and flexible fashion, to easily accommodate new features. Currently, it implements an array of state-based solvers that addresses the state explosion problem through powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. A simple, yet representative, case study is adopted, to present the tool and to show the feasibility of the supported methods, in particular from memory consumption point of view.Source: TOSME Workshop - Tools for Stochastic Modelling and Evaluation (performance, dependability, security and verification), Online workshop, 12/11/2021
See at:
ISTI Repository | www.performance2021.deib.polimi.it | CNR ExploRA
2020
Journal article
Open Access
Computing performability measures in Markov chains by means of matrix functions
Masetti G., Robol L.
We discuss the efficient computation of performance, reliability, and availability measures for Markov chains; these metrics - and the ones obtained by combining them, are often called performability measures. We show that this computational problem can be recasted as the evaluation of a bilinear form induced by appropriate matrix functions, and thus solved by leveraging the fast methods available for this task.Source: Journal of computational and applied mathematics 368 (2020). doi:10.1016/j.cam.2019.112534
DOI: 10.1016/j.cam.2019.112534
DOI: 10.48550/arxiv.1803.06322
Metrics:
Masetti G., Robol L.
We discuss the efficient computation of performance, reliability, and availability measures for Markov chains; these metrics - and the ones obtained by combining them, are often called performability measures. We show that this computational problem can be recasted as the evaluation of a bilinear form induced by appropriate matrix functions, and thus solved by leveraging the fast methods available for this task.Source: Journal of computational and applied mathematics 368 (2020). doi:10.1016/j.cam.2019.112534
DOI: 10.1016/j.cam.2019.112534
DOI: 10.48550/arxiv.1803.06322
Metrics:
See at:
arXiv.org e-Print Archive | Journal of Computational and Applied Mathematics | ISTI Repository | ISTI Repository | Journal of Computational and Applied Mathematics | doi.org | www.sciencedirect.com | CNR ExploRA
2020
Conference article
Open Access
Analyzing Forward Robustness of Feedforward Deep Neural Networks with LeakyReLU Activation Function Through Symbolic Propagation
Masetti G., Di Giandomenico F.
FeedForward Deep Neural Networks (DNNs) robustness is a relevant property to study, since it allows to establish whether the classification performed by DNNs is vulnerable to small perturbations in the provided input, and several verification approaches have been developed to assess such robustness degree. Recently, an approach has been introduced to evaluate forward robustness, based on symbolic computations and designed for the ReLU activation function. In this paper, a generalization of such a symbolic approach for the widely adopted LeakyReLU activation function is developed. A preliminary numerical campaign, briefly discussed in the paper, shows interesting results.Source: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 460–474, 14/09/2020
DOI: 10.1007/978-3-030-65965-3_31
Metrics:
Masetti G., Di Giandomenico F.
FeedForward Deep Neural Networks (DNNs) robustness is a relevant property to study, since it allows to establish whether the classification performed by DNNs is vulnerable to small perturbations in the provided input, and several verification approaches have been developed to assess such robustness degree. Recently, an approach has been introduced to evaluate forward robustness, based on symbolic computations and designed for the ReLU activation function. In this paper, a generalization of such a symbolic approach for the widely adopted LeakyReLU activation function is developed. A preliminary numerical campaign, briefly discussed in the paper, shows interesting results.Source: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 460–474, 14/09/2020
DOI: 10.1007/978-3-030-65965-3_31
Metrics:
See at:
ISTI Repository | doi.org | link.springer.com | CNR ExploRA
2020
Conference article
Open Access
Trading dependability and energy consumption in critical infrastructures: Focus on the rail switch heating system
Chiaradonna S., Di Giandomenico F., Masetti G.
Traditionally, critical infrastructures demand for high dependability, being the services they provide essential to human beings and the society at large. However, more recent attention to cautious usage of energy resources is changing this vision and calls for solutions accounting for appropriate multi-requirements combinations when developing a critical infrastructure. In such a context, analysis supports able to assist the designer in envisioning a satisfactory trade-off among the multi-requirements for the system at hand are highly helpful. In this paper, the focus is on the railway sector and the contribution is a stochastic model-based analysis framework to quantitatively assess trade-offs between dependability indicators and electrical energy consumption incurred by the rail switch heating system.Moving from a preliminary study that concentrated on energy consumption only, the analysis framework has been extended to become a solid support to devise appropriate tuning of the heating policy that guarantees satisfactory trade-offs between dependability and energy consumption. An evaluation campaign in a variety of climate scenarios demonstrates the feasibility and utility of the developed framework.Source: 25th IEEE Pacific Rim International Symposium on Dependable Computing, pp. 150–159, Perth, Australia, 01/12/2021
DOI: 10.1109/prdc50213.2020.00026
Metrics:
Chiaradonna S., Di Giandomenico F., Masetti G.
Traditionally, critical infrastructures demand for high dependability, being the services they provide essential to human beings and the society at large. However, more recent attention to cautious usage of energy resources is changing this vision and calls for solutions accounting for appropriate multi-requirements combinations when developing a critical infrastructure. In such a context, analysis supports able to assist the designer in envisioning a satisfactory trade-off among the multi-requirements for the system at hand are highly helpful. In this paper, the focus is on the railway sector and the contribution is a stochastic model-based analysis framework to quantitatively assess trade-offs between dependability indicators and electrical energy consumption incurred by the rail switch heating system.Moving from a preliminary study that concentrated on energy consumption only, the analysis framework has been extended to become a solid support to devise appropriate tuning of the heating policy that guarantees satisfactory trade-offs between dependability and energy consumption. An evaluation campaign in a variety of climate scenarios demonstrates the feasibility and utility of the developed framework.Source: 25th IEEE Pacific Rim International Symposium on Dependable Computing, pp. 150–159, Perth, Australia, 01/12/2021
DOI: 10.1109/prdc50213.2020.00026
Metrics:
See at:
ISTI Repository | doi.org | ieeexplore.ieee.org | CNR ExploRA
2020
Contribution to conference
Open Access
Enhancing sustainability of the railway infrastructure through efficient energy management policies
Chiaradonna S., Masetti G., Di Giandomenico F., Righetti F., Vallati C.
Railway is currently envisioned as the most promising transportation system for both people and freight to reduce atmospheric emission and combat climate change. In this context, ensuring the energy efficiency of the railway systems is paramount in order to sustain their future expandability with minimum carbon footprint. Recent advancements in computing and communication technologies are expected to play a significant role to enable novel integrated control and management strategies in which heterogeneous data is exploited to noticeably increase energy efficiency. In this paper we focus on exploiting the convergence of heterogeneous information to improve energy efficiency of railway systems, in particular on the heating system for the railroad switches, one of the major energy intensive components. To this aim, we define new policies to efficiently manage the heating of these switches exploiting also external information such as weather and forecast data. In order to assess the performance of each strategy, a stochastic model representing the structure and operation of the railroad switch heating system and environmental conditions (both weather profiles and specific failure events) has been developed and exercised in a variety of representative scenarios. The obtained results allow to understand both strengths and limitations of each energy management policy, and serves as a useful support to make the choice of the best technique to employ to save on energy consumption, given the system conditions at hand.Source: 11th International Green and Sustainable Computing Conference, Virtual Conference, 19/10/2020, 22/10/2020
Chiaradonna S., Masetti G., Di Giandomenico F., Righetti F., Vallati C.
Railway is currently envisioned as the most promising transportation system for both people and freight to reduce atmospheric emission and combat climate change. In this context, ensuring the energy efficiency of the railway systems is paramount in order to sustain their future expandability with minimum carbon footprint. Recent advancements in computing and communication technologies are expected to play a significant role to enable novel integrated control and management strategies in which heterogeneous data is exploited to noticeably increase energy efficiency. In this paper we focus on exploiting the convergence of heterogeneous information to improve energy efficiency of railway systems, in particular on the heating system for the railroad switches, one of the major energy intensive components. To this aim, we define new policies to efficiently manage the heating of these switches exploiting also external information such as weather and forecast data. In order to assess the performance of each strategy, a stochastic model representing the structure and operation of the railroad switch heating system and environmental conditions (both weather profiles and specific failure events) has been developed and exercised in a variety of representative scenarios. The obtained results allow to understand both strengths and limitations of each energy management policy, and serves as a useful support to make the choice of the best technique to employ to save on energy consumption, given the system conditions at hand.Source: 11th International Green and Sustainable Computing Conference, Virtual Conference, 19/10/2020, 22/10/2020
See at:
ISTI Repository | CNR ExploRA
2020
Conference article
Open Access
Failure management strategies for IoT-based railways systems
Righetti F., Vallati C., Anastasi G., Masetti G., Di Giandomenico F.
Railways monitoring and control are currently performed by different heterogeneous vertical systems working in isolation without or with limited cooperation among them. Such configuration, widely adopted in practical deployments today, is in contrast with the integrated vision of systems that are at the foundation of the smart-city concept. In order to overcome the current fractured ecosystem that monitors and controls railways functionalities, the adoption of a novel integrated approach is mandatory to create an all-in-one railway system. To this aim, new IoT-based communication technologies, like wireless or Power Line Communication technologies, are considered the main enablers to integrate in a very rapid and easy manner existing vertical systems. In this work, we analyse the architecture of future railways systems based on a mix of wireless and Power Line Communication technologies. In our analysis, we aim at studying possible failure management strategies on rail-road switches to improve the level of reliability, crucial requirement for systems that demand maximum resiliency as they manage a critical function of the infrastructure. In particular, we propose a set of solutions aimed at detecting and handling network and sensor failures to ensure continuity in the execution of the basic control functions. The proposed approach is evaluated by means of simulations and demonstrated to be effective in ensuring a good level of performance even when failures occur.Source: 2020 IEEE International Conference on Smart Computing, pp. 386–391, Bologna, 14-17/09/2020
DOI: 10.1109/smartcomp50058.2020.00082
Metrics:
Righetti F., Vallati C., Anastasi G., Masetti G., Di Giandomenico F.
Railways monitoring and control are currently performed by different heterogeneous vertical systems working in isolation without or with limited cooperation among them. Such configuration, widely adopted in practical deployments today, is in contrast with the integrated vision of systems that are at the foundation of the smart-city concept. In order to overcome the current fractured ecosystem that monitors and controls railways functionalities, the adoption of a novel integrated approach is mandatory to create an all-in-one railway system. To this aim, new IoT-based communication technologies, like wireless or Power Line Communication technologies, are considered the main enablers to integrate in a very rapid and easy manner existing vertical systems. In this work, we analyse the architecture of future railways systems based on a mix of wireless and Power Line Communication technologies. In our analysis, we aim at studying possible failure management strategies on rail-road switches to improve the level of reliability, crucial requirement for systems that demand maximum resiliency as they manage a critical function of the infrastructure. In particular, we propose a set of solutions aimed at detecting and handling network and sensor failures to ensure continuity in the execution of the basic control functions. The proposed approach is evaluated by means of simulations and demonstrated to be effective in ensuring a good level of performance even when failures occur.Source: 2020 IEEE International Conference on Smart Computing, pp. 386–391, Bologna, 14-17/09/2020
DOI: 10.1109/smartcomp50058.2020.00082
Metrics:
See at:
ISTI Repository | doi.org | ieeexplore.ieee.org | CNR ExploRA
2020
Journal article
Open Access
Analysis of violin combination tones and their contribution to Tartini's third tone
Caselli G., Cecchi G., Malacarne M., Masetti G.
It is widely accepted that the famous Tartini's third tone, i.e., the appearance of an additional third tone of lower frequency when playing a dyad on the violin, is a subjective phenomenon generated by the listener's cochlear nonlinearity. However, the recent demonstration that additional tones of audible amplitude can also be generated by the violin itself during playing of a dyad (violin combination tones), raises the question if these tones might have influenced Tartini's third sound perception. The experiments reported here were made to ascertain this possibility. To this end, following Tartini experiments, several dyads played by either one violin or two violins playing one note of the dyad each, were recorded. The analysis of the spectra shows that violin combination tones are present in all the dyads investigated, but exclusively when the dyad is played by a single violin and not when the same dyad is played by two violins. Tartini found the third tones to be the same in both conditions, which means that violin combination tones in his experiments were either absent or too small to affect the perception of the subjective third tones arising from cochlear distortion.Source: Savart journal 1 (2020).
Caselli G., Cecchi G., Malacarne M., Masetti G.
It is widely accepted that the famous Tartini's third tone, i.e., the appearance of an additional third tone of lower frequency when playing a dyad on the violin, is a subjective phenomenon generated by the listener's cochlear nonlinearity. However, the recent demonstration that additional tones of audible amplitude can also be generated by the violin itself during playing of a dyad (violin combination tones), raises the question if these tones might have influenced Tartini's third sound perception. The experiments reported here were made to ascertain this possibility. To this end, following Tartini experiments, several dyads played by either one violin or two violins playing one note of the dyad each, were recorded. The analysis of the spectra shows that violin combination tones are present in all the dyads investigated, but exclusively when the dyad is played by a single violin and not when the same dyad is played by two violins. Tartini found the third tones to be the same in both conditions, which means that violin combination tones in his experiments were either absent or too small to affect the perception of the subjective third tones arising from cochlear distortion.Source: Savart journal 1 (2020).
See at:
ISTI Repository | www.savartjournal.org | CNR ExploRA
2019
Report
Open Access
Tensor methods for the computation of MTTF in large systems of loosely interconnected components
Masetti G., Robol L.
We are concerned with the computation of the mean-time-to-failure(MTTF) for a large system of loosely interconnected components, mod-eled as continuous time Markov chains. In particular, we show that split-ting the local and synchronization transitions of the smaller subsystemsallows to formulate an algorithm for the computation of the MTTF whichis proven to be linearly convergent. Then, we show how to modify themethod to make it quadratically convergent, thus overcoming the difficul-ties for problems with convergent rate close to1.In addition, it is shown that this decoupling of local and synchroniza-tion transitions allows to easily represent all the matrices and vectors in-volved in the method in the tensor-train (TT) format -- and we providenumerical evidence showing that this allows to treat large problems withup to billions of states -- which would otherwise be unfeasible.Source: ISTI Technical reports, 2019
Masetti G., Robol L.
We are concerned with the computation of the mean-time-to-failure(MTTF) for a large system of loosely interconnected components, mod-eled as continuous time Markov chains. In particular, we show that split-ting the local and synchronization transitions of the smaller subsystemsallows to formulate an algorithm for the computation of the MTTF whichis proven to be linearly convergent. Then, we show how to modify themethod to make it quadratically convergent, thus overcoming the difficul-ties for problems with convergent rate close to1.In addition, it is shown that this decoupling of local and synchroniza-tion transitions allows to easily represent all the matrices and vectors in-volved in the method in the tensor-train (TT) format -- and we providenumerical evidence showing that this allows to treat large problems withup to billions of states -- which would otherwise be unfeasible.Source: ISTI Technical reports, 2019
See at:
dcl.isti.cnr.it | ISTI Repository | CNR ExploRA
2019
Conference article
Open Access
Distinguishing Violinists and Pianists Based on Their Brain Signals
Coro G., Masetti G., Bonhoeffer P., Betcher M.
Many studies in neuropsychology have highlighted that expert musicians, who started learning music in childhood, present structural differences in their brains with respect to non-musicians. This indicates that early music learning affects the development of the brain. Also, musicians' neuronal activity is different depending on the played instrument and on the expertise. This difference can be analysed by processing electroencephalographic (EEG) signals through Artificial Intelligence models. This paper explores the feasibility to build an automatic model that distinguishes violinists from pianists based only on their brain signals. To this aim, EEG signals of violinists and pianists are recorded while they play classical music pieces and an Artificial Neural Network is trained through a cloud computing platform to build a binary classifier of segments of these signals. Our model has the best classification performance on 20 seconds EEG segments, but this performance depends on the involved musicians' expertise. Also, the brain signals of a cellist are demonstrated to be more similar to violinists' signals than to pianists' signals. In summary, this paper demonstrates that distinctive information is present in the two types of musicians' brain signals, and that this information can be detected even by an automatic model working with a basic EEG equipment.Source: ICANN 2019: Theoretical Neural Computation 28th International Conference on Artificial Neural Networks, pp. 123–137, Monaco di Baviera, 17-19/10/2019
DOI: 10.1007/978-3-030-30487-4_11
Metrics:
Coro G., Masetti G., Bonhoeffer P., Betcher M.
Many studies in neuropsychology have highlighted that expert musicians, who started learning music in childhood, present structural differences in their brains with respect to non-musicians. This indicates that early music learning affects the development of the brain. Also, musicians' neuronal activity is different depending on the played instrument and on the expertise. This difference can be analysed by processing electroencephalographic (EEG) signals through Artificial Intelligence models. This paper explores the feasibility to build an automatic model that distinguishes violinists from pianists based only on their brain signals. To this aim, EEG signals of violinists and pianists are recorded while they play classical music pieces and an Artificial Neural Network is trained through a cloud computing platform to build a binary classifier of segments of these signals. Our model has the best classification performance on 20 seconds EEG segments, but this performance depends on the involved musicians' expertise. Also, the brain signals of a cellist are demonstrated to be more similar to violinists' signals than to pianists' signals. In summary, this paper demonstrates that distinctive information is present in the two types of musicians' brain signals, and that this information can be detected even by an automatic model working with a basic EEG equipment.Source: ICANN 2019: Theoretical Neural Computation 28th International Conference on Artificial Neural Networks, pp. 123–137, Monaco di Baviera, 17-19/10/2019
DOI: 10.1007/978-3-030-30487-4_11
Metrics:
See at:
ISTI Repository | doi.org | link.springer.com | CNR ExploRA
2019
Journal article
Open Access
On Extending and Comparing Newton-Raphson Variants for Solving Power-Flow Equations
Dutto S., Masetti G., Chiaradonna S., Di Giandomenico F.
This paper focuses on power-flow equations solutions, based on the Newton-Raphson method. Two major contributions are offered. First, the definition of novel solution variants, resorting to Wirtinger calculus, is attempted. The obtained developments, although original in their formulation, led to already known variants. Despite the impaired originality of the obtained solution, there are significant lessons learned from such an effort. The second major contribution consists of a deep comparison analysis of existing solution strategies, based on complex and real variables, and the Wirtinger based ones, all properly reformulated to allow direct comparison with each other. The goal is to investigate strengths and weaknesses of the addressed techniques in terms of computational effort and convergence rate, which are the most relevant aspects to consider while choosing the approach to employ to solve power-flow equations for a specific power system under study.Source: IEEE transactions on power systems 34 (2019): 2577–2587. doi:10.1109/TPWRS.2019.2897640
DOI: 10.1109/tpwrs.2019.2897640
Metrics:
Dutto S., Masetti G., Chiaradonna S., Di Giandomenico F.
This paper focuses on power-flow equations solutions, based on the Newton-Raphson method. Two major contributions are offered. First, the definition of novel solution variants, resorting to Wirtinger calculus, is attempted. The obtained developments, although original in their formulation, led to already known variants. Despite the impaired originality of the obtained solution, there are significant lessons learned from such an effort. The second major contribution consists of a deep comparison analysis of existing solution strategies, based on complex and real variables, and the Wirtinger based ones, all properly reformulated to allow direct comparison with each other. The goal is to investigate strengths and weaknesses of the addressed techniques in terms of computational effort and convergence rate, which are the most relevant aspects to consider while choosing the approach to employ to solve power-flow equations for a specific power system under study.Source: IEEE transactions on power systems 34 (2019): 2577–2587. doi:10.1109/TPWRS.2019.2897640
DOI: 10.1109/tpwrs.2019.2897640
Metrics:
See at:
ISTI Repository | IEEE Transactions on Power Systems | ieeexplore.ieee.org | CNR ExploRA
2019
Conference article
Open Access
Stochastic modeling and evaluation of large interdependent composed models through Kronecker algebra and exponential sums
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
The KAES methodology for efficient evaluation of dependability-related properties is proposed. KAES targets systems representable by Stochastic Petri Nets-based models, composed by a large number of submodels where interconnections are managed through synchronization at action level. The core of KAES is a new numerical solution of the underlying CTMC process, based on powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. Specifically, advancing on existing literature, KAES addresses efficient evaluation of the Mean-Time-To-Absorption in CTMC with absorbing states, exploiting the basic idea to further pursue the symbolic representation of the elements involved in the evaluation process, so to better cope with the problem of state explosion. As a result, computation efficiency is improved, especially when the submodels are loosely interconnected and have small number of states. An instrumental case study is adopted, to show the feasibility of KAES, in particular from memory consumption point of view.Source: The 40th International Conference on Application and Theory of Petri Nets and Concurrency, pp. 47–66, Berlin, 23-28/06/2019
DOI: 10.1007/978-3-030-21571-2_3
Metrics:
Masetti G., Robol L., Chiaradonna S., Di Giandomenico F.
The KAES methodology for efficient evaluation of dependability-related properties is proposed. KAES targets systems representable by Stochastic Petri Nets-based models, composed by a large number of submodels where interconnections are managed through synchronization at action level. The core of KAES is a new numerical solution of the underlying CTMC process, based on powerful mathematical techniques, including Kronecker algebra, Tensor Trains and Exponential Sums. Specifically, advancing on existing literature, KAES addresses efficient evaluation of the Mean-Time-To-Absorption in CTMC with absorbing states, exploiting the basic idea to further pursue the symbolic representation of the elements involved in the evaluation process, so to better cope with the problem of state explosion. As a result, computation efficiency is improved, especially when the submodels are loosely interconnected and have small number of states. An instrumental case study is adopted, to show the feasibility of KAES, in particular from memory consumption point of view.Source: The 40th International Conference on Application and Theory of Petri Nets and Concurrency, pp. 47–66, Berlin, 23-28/06/2019
DOI: 10.1007/978-3-030-21571-2_3
Metrics:
See at:
arpi.unipi.it | link.springer.com | ISTI Repository | doi.org | CNR ExploRA