1,721,017 research outputs found
Resilience of Deep Learning Applications: Where We are and Where We Want to Go
No full textDeep Learning (DL) [1] is currently one of the most intensively and widely used predictive models in the field of machine learning. DL has proven to give very good results for many complex tasks and applications, such as object recognition in images/videos, natural language processing, robotics, aerospace, smart healthc are, and autonomous driving. Nowa-days, there is intense activity in designing custom Artificial Intelligence (AI) hardware accelerators to support the energy-hungry data movement, speed of computation, and memory resources that DL requires to realize its full potential [2]. Furthermore, there is an incentive to migrate AI from cloud to edge devices, i.e., Internet-of- Things devices, to address data confidentiality issues and bandwidth limitations, and also to alleviate the communication latency, especially for real-time safety-critical decisions, e.g., in autonomous driving
Board-level functional fault diagnosis using data mining
No full textThis paper presents an approach for performing functional diagnosis of complex systems by means of data mining. The technique allows to derive a set of rules from a functional model of the system for efficiently driving the diagnosis procedure towards the identification of the most promising faulty candidate. The approach is adopted within an incremental method, to limit the number of tests to be performed, thus reducing costs and effort
Aiming for an Online/ Onsite Format and Finally Moving to Online Only
Full text linkBased on the past two years of experience as an online event due to the Covid-19 outbreak, DATE 2022 has been planned to cope with the uncertainty of the situation, having two days onsite, in Antwerp, Belgium, from March 14 to 15, 2022, to be followed by an online program until March 23, 2022. A different organization and format had been adopted to engage the DATE community, with a two-day rich program and numerous talks and panels to gather again, in person. The scientific program online had been completed by live panels and presentations to offer a valuable virtual experience. Unfortunately, for the third year in a row, DATE moved to a completely virtual event, which nevertheless attracted a broad audience.status: Published onlin
Emergent Semantics and Cooperation in Multi-Knowledge Environments: the ESTEEM Architecture
No full textIn the present global society, information has to be exchangeable in open and dynamic environments, where interacting peers do not necessarily share a common understanding of the world at hand, and do not have a complete picture of the context where the interaction occurs. In this paper, we present the Esteem approach and the related peer architecture for emergent semantics in dynamic and multi-knowledge environments. In Esteem, semantic communities are built around declared interests in the form of manifesto ontologies, and their autonomous nature is preserved by allowing a shared semantics to naturally emerge from peer interactions
Error Modeling for Image Processing Filters accelerated onto SRAM-based FPGAs
Full text linkImage processing is today employed in a variety of application fields, including safety-and mission-critical ones. In these scenarios it is vital to carefully analyse the reliability of the designed system before deployment and, if necessary, to adopt specific hardening techniques. Two are the techniques generally employed: circuit-level fault injection and application-level functional error simulation. In this paper we present a set of functional error models specific for a number of convolution-based filters that are the basic building blocks for a wide range of image processing applications. The presented error models, derived through a number of circuit-level fault injection experiments, may be integrated into application-level functional error simulators, bridging the gap between the two strategies. The presented error models are the first step towards combining the accuracy of fault injection and the flexibility of error simulation into a widely adopted reliability analysis tool
Analyzing the Reliability of Alternative Convolution Implementations for Deep Learning Applications
Full text linkConvolution represents the core of Deep Learning (DL) applications, enabling the automatic extraction of features from raw input data. Several implementations of the convolution have been proposed. The impact of these different implementations on the performance of DL applications has been studied. However, no specific reliability-related analysis has been carried out. In this paper, we apply the CLASSES cross-layer reliability analysis methodology for an in-depth study aimed at: i) analyzing and characterizing the effects of Single Event Upsets occurring in Graphics Processing Units while executing the convolution operators; and ii) identifying whether a convolution implementation is more robust than others. The outcomes can then be exploited to tailor better hardening schemes for DL applications to improve reliability and reduce overhead
- …
