Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

On Neural Networks Redundancy and Diversity for Their Use in Safety-Critical Systems

  • Brando, Axel
  • Serra, Isabel
  • Mezzetti, Enrico|||0000-0002-1886-2931
  • Cazorla Almeida, Francisco Javier
  • Perez Cerrolaza, Jon
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
An increasing number of critical functionalities integrated in embedded critical systems rely on deep learning (DL) technology. This article summarizes certain key aspects of DL’s intrinsic stochastic and training-data-dependent nature that are at odds with current domain-specific functional safety standards. We exemplify how redundancy and diversity of neural networks can help to reconcile DL technology and functional safety requirements., The research leading to these results has received funding from the European Research Council (ERC) grant agreement No. 772773 (SuPerCom), the Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num.101069595, and the Spanish Ministry of Science and Innovation under grant PID2019-107255GBC21/AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Assessing the use of NVIDIA multi-instance GPU in the automotive domain

  • Barrera Herrera, Javier Enrique|||0000-0003-1469-2514
  • Kosmidis, Leonidas
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
  • Cazorla Almeida, Francisco Javier
The search for resource isolation and segregation is relentless to increase execution time determinism (ETD) as an essential feature for a successful certification process in embedded critical domains like automotive. In those domains, the advent of advanced software-controlled features requires unprecedented computing performance that calls for the use of acceleration hardware, with GPUs having a prominent position. The latest NVIDIA GPUs generations (Ampere, Hopper, Blackwell) feature MIG (Multi-Instance GPU), a mechanism that allows partitioning the GPU into fully isolated GPU instances, each with its own memory, cache, and computing resources. Despite the clear benefits of MIG on ETD, the latest NVIDIA automotive GPUs do not implement it. In this work, we first empirically analyze the benefits of MIG in a non-automotive GPU showing the main traits in its use to improve ETD. And second, we identify the potential reasons precluding the deployment of MIG for automotive GPUs: automotive market specific needs, and the difference between GPU and memory technologies used in high-performance GPUs, which implement MIG, and automotive GPUs that lack it., This work was supported in part by MICIU/AEI/10.13039/501100011033 under Project PLEC2023-010240; in part by the European Union’s HE Research and Innovation Programme under Grant 101069595; and in part by Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR) under Grant 2021 SGR 00637. The work of Javier Barrera was supported by the “Generación D” initiative, Red.es, Ministerio para la Transformación Digital y de la Función Pública, for talent attraction, funded by the European Union NextGenerationEU funds, through PRTR under Grant C005/24-ED CV1. The work of Leonidas Kosmidis was supported by MICIU/AEI/10.13039/ 501100011033 under Grant IJC-2020- 045931-I., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Semantic diverse DMR and TMR for high-integrity AI-based function efficiency

  • Caro Roca, Martí
  • Brando Guillaumes, Axel
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
Dual (DMR) and Triple Modular Redundancy (TMR), often with some form of diversity, are used in safety-critical systems to realize those functionalities at the highest integrity level providing fault detection and/or tolerance capabilities. Redundant executions are intended to provide bit-level identical results and, upon any mismatch, an error is assumed and recovery actions taken as needed. In this paper, we note that many emerging AI-based functionalities are intrinsically stochastic (e.g., camera-based object detection), and hence, their correctness must be judged semantically, with room for variations across correct outcomes (e.g., confidence must be above a given threshold, but how much it exceeds the threshold is irrelevant). Building on this observation, we propose strategies to create DMR and TMR implementations of AI-based functionalities that bring, not only fault tolerance against random hardware faults, but also against AI model inaccuracies. Those strategies, which can be realized with software-only means and ported to virtually any computing platform, build on input data modifications affecting the inference computations, but not the expected semantic output (e.g., introducing some controlled changes in the input data). Moreover, we provide our solution in the form of an open-source tool for image and video processing aimed at facilitating the reproducibility of our evaluation results, and enabling others to use it and conduct further research on input transformations., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595. This work is also part of the project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033. Authors appreciate the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya. M. Caro has received funding under grant PRE2022-102440 funded by MICIU/AEI/10.13039/501100011033 and, by ESF+., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

On Neural Networks Redundancy and Diversity for Their Use in Safety-Critical Systems

  • Brando, Axel
  • Serra, Isabel
  • Mezzetti, Enrico
  • Cazorla Almeida, Francisco Javier
  • Perez Cerrolaza, Jon
  • Abella Ferrer, Jaume
An increasing number of critical functionalities integrated in embedded critical systems rely on deep learning (DL) technology. This article summarizes certain key aspects of DL’s intrinsic stochastic and training-data-dependent nature that are at odds with current domain-specific functional safety standards. We exemplify how redundancy and diversity of neural networks can help to reconcile DL technology and functional safety requirements., The research leading to these results has received funding from the European Research Council (ERC) grant agreement No. 772773 (SuPerCom), the Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num.101069595, and the Spanish Ministry of Science and Innovation under grant PID2019-107255GBC21/AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Semantic diverse DMR and TMR for high-integrity AI-based function efficiency

  • Caro Roca, Martí
  • Brando Guillaumes, Axel
  • Abella Ferrer, Jaume
Dual (DMR) and Triple Modular Redundancy (TMR), often with some form of diversity, are used in safety-critical systems to realize those functionalities at the highest integrity level providing fault detection and/or tolerance capabilities. Redundant executions are intended to provide bit-level identical results and, upon any mismatch, an error is assumed and recovery actions taken as needed. In this paper, we note that many emerging AI-based functionalities are intrinsically stochastic (e.g., camera-based object detection), and hence, their correctness must be judged semantically, with room for variations across correct outcomes (e.g., confidence must be above a given threshold, but how much it exceeds the threshold is irrelevant). Building on this observation, we propose strategies to create DMR and TMR implementations of AI-based functionalities that bring, not only fault tolerance against random hardware faults, but also against AI model inaccuracies. Those strategies, which can be realized with software-only means and ported to virtually any computing platform, build on input data modifications affecting the inference computations, but not the expected semantic output (e.g., introducing some controlled changes in the input data). Moreover, we provide our solution in the form of an open-source tool for image and video processing aimed at facilitating the reproducibility of our evaluation results, and enabling others to use it and conduct further research on input transformations., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595. This work is also part of the project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033. Authors appreciate the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya. M. Caro has received funding under grant PRE2022-102440 funded by MICIU/AEI/10.13039/501100011033 and, by ESF+., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Assessing the use of NVIDIA multi-instance GPU in the automotive domain

  • Barrera Herrera, Javier Enrique
  • Kosmidis, Leonidas
  • Abella Ferrer, Jaume
  • Cazorla Almeida, Francisco Javier
The search for resource isolation and segregation is relentless to increase execution time determinism (ETD) as an essential feature for a successful certification process in embedded critical domains like automotive. In those domains, the advent of advanced software-controlled features requires unprecedented computing performance that calls for the use of acceleration hardware, with GPUs having a prominent position. The latest NVIDIA GPUs generations (Ampere, Hopper, Blackwell) feature MIG (Multi-Instance GPU), a mechanism that allows partitioning the GPU into fully isolated GPU instances, each with its own memory, cache, and computing resources. Despite the clear benefits of MIG on ETD, the latest NVIDIA automotive GPUs do not implement it. In this work, we first empirically analyze the benefits of MIG in a non-automotive GPU showing the main traits in its use to improve ETD. And second, we identify the potential reasons precluding the deployment of MIG for automotive GPUs: automotive market specific needs, and the difference between GPU and memory technologies used in high-performance GPUs, which implement MIG, and automotive GPUs that lack it., This work was supported in part by MICIU/AEI/10.13039/501100011033 under Project PLEC2023-010240; in part by the European Union’s HE Research and Innovation Programme under Grant 101069595; and in part by Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR) under Grant 2021 SGR 00637. The work of Javier Barrera was supported by the “Generación D” initiative, Red.es, Ministerio para la Transformación Digital y de la Función Pública, for talent attraction, funded by the European Union NextGenerationEU funds, through PRTR under Grant C005/24-ED CV1. The work of Leonidas Kosmidis was supported by MICIU/AEI/10.13039/ 501100011033 under Grant IJC-2020- 045931-I., Peer Reviewed, Postprint (author's final draft)