Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- research-articleJuly 2024JUST ACCEPTED
M2SKD: Multi-to-Single Knowledge Distillation of Real-Time Epileptic Seizure Detection for Low-Power Wearable Systems
ACM Transactions on Intelligent Systems and Technology (TIST), Just Accepted https://doi.org/10.1145/3675402Integrating low-power wearable systems into routine health monitoring is an ongoing challenge. Recent advances in the computation capabilities of wearables make it possible to target complex scenarios by exploiting multiple biosignals and using high-...
- extended-abstractJuly 2024
Performance evaluation of acceleration of convolutional layers on OpenEdgeCGRA
- Nicolò Carpentieri,
- Juan Sapriza,
- Davide Schiavone,
- Daniele Jahier Pagliari,
- David Atienza,
- Maurizio Martina,
- Alessio Burrello
CF '24 Companion: Proceedings of the 21st ACM International Conference on Computing Frontiers: Workshops and Special SessionsMay 2024, Pages 67–70https://doi.org/10.1145/3637543.3652875Recently, efficiently deploying deep learning solutions on the edge has received increasing attention. New platforms are emerging to support the increasing demand for flexibility and high performance. In this work, we explore the efficient mapping of ...
- research-articleMay 2024JUST ACCEPTED
SAT-based Exact Modulo Scheduling Mapping for Resource-Constrained CGRAs
- Cristian Tirelli,
- Juan Sapriza,
- Rubén Rodríguez Álvarez,
- Lorenzo Ferretti,
- Benoît Denkinger,
- Giovanni Ansaloni,
- José Miranda Calero,
- David Atienza,
- Laura Pozzi
ACM Journal on Emerging Technologies in Computing Systems (JETC), Just Accepted https://doi.org/10.1145/3663675Coarse-Grain Reconfigurable Arrays (CGRAs) represent emerging low-power architectures designed to accelerate Compute-Intensive Loops (CILs). The effectiveness of CGRAs in providing acceleration relies on the quality of mapping: how efficiently the CIL is ...
- research-articleApril 2024JUST ACCEPTED
Intermediate Address Space: virtual memory optimization of heterogeneous architectures for cache-resident workloads
ACM Transactions on Architecture and Code Optimization (TACO), Just Accepted https://doi.org/10.1145/3659207The increasing demand for computing power and the emergence of heterogeneous computing architectures have driven the exploration of innovative techniques to address current limitations in both the compute and memory subsystems. One such solution is the ...
- research-articleApril 2024
Combining general and personal models for epilepsy detection with hyperdimensional computing
Artificial Intelligence in Medicine (AIIM), Volume 148, Issue CFeb 2024https://doi.org/10.1016/j.artmed.2023.102754AbstractEpilepsy is a highly prevalent chronic neurological disorder with great negative impact on patients’ daily lives. Despite this there is still no adequate technological support to enable epilepsy detection and continuous outpatient monitoring in ...
Highlights- Hyperdimensional computing (HDC) models have various advantages for epilepsy detection, for example easy and interpretable comparison of seizure and non-seizure models.
- General models can be created from individual personal models, but ...
-
- research-articleJanuary 2024
Special Session: Challenges and Opportunities for Sustainable Multi-Scale Computing Systems
CODES/ISSS '23 Companion: Proceedings of the 2023 International Conference on Hardware/Software Codesign and System SynthesisSeptember 2023, Pages 24–27https://doi.org/10.1145/3607888.3608961Multi-Scale computing systems aim at bringing the computing as close as possible to the data sources, to optimize both computation and networking. These systems are composed of at least three computing layers: the terminal layer, the edge layer, and the ...
- research-articleNovember 2023
A semi-supervised algorithm for improving the consistency of crowdsourced datasets: The COVID-19 case study on respiratory disorder classification
Computer Methods and Programs in Biomedicine (CBIO), Volume 241, Issue CNov 2023https://doi.org/10.1016/j.cmpb.2023.107743AbstractBackground and Objective: Cough audio signal classification is a potentially useful tool in screening for respiratory disorders, such as COVID-19. Since it is dangerous to collect data from patients with contagious diseases, many research teams ...
Highlights
- We apply semi-supervised learning to COVID-19 screening from cough sounds.
- This explainable ML approach improves label consistency and class separability.
- Our method can be used to overcome human label scarcity and expert ...
- research-articleOctober 2023
Event-based sampled ECG morphology reconstruction through self-similarity
Computer Methods and Programs in Biomedicine (CBIO), Volume 240, Issue COct 2023https://doi.org/10.1016/j.cmpb.2023.107712Highlights- Event-based (EB) sampling provides a lossy representation of a signal.
- Self-similarity is an electrocardiogram property that allows signal representation through templates.
- Locally representative templates can be updated to remain ...
Background and Objective: Event-based analog-to-digital converters allow for sparse bio-signal acquisition, enabling local sub-Nyquist sampling frequency. However, aggressive event selection can cause the loss of important bio-markers, not ...
- research-articleSeptember 2023
Overflow-free Compute Memories for Edge AI Acceleration
ACM Transactions on Embedded Computing Systems (TECS), Volume 22, Issue 5sArticle No.: 121, Pages 1–23https://doi.org/10.1145/3609387Compute memories are memory arrays augmented with dedicated logic to support arithmetic. They support the efficient execution of data-centric computing patterns, such as those characterizing Artificial Intelligence (AI) algorithms. These architectures can ...
- research-articleSeptember 2023
Acceleration of Control Intensive Applications on Coarse-Grained Reconfigurable Arrays for Embedded Systems
IEEE Transactions on Computers (ITCO), Volume 72, Issue 9Sept. 2023, Pages 2548–2560https://doi.org/10.1109/TC.2023.3257504Embedded systems confront two opposite goals: low-power operation and high performance. The current trend to reach these goals is toward heterogeneous platforms, including multi-core architectures with heterogeneous cores and hardware accelerators. The ...
- extended-abstractAugust 2023
An Open-Hardware Coarse-Grained Reconfigurable Array for Edge Computing
- Rubén Rodríguez Álvarez,
- Benoît Denkinger,
- Juan Sapriza,
- José Miranda Calero,
- Giovanni Ansaloni,
- David Atienza Alonso
CF '23: Proceedings of the 20th ACM International Conference on Computing FrontiersMay 2023, Pages 391–392https://doi.org/10.1145/3587135.3591437In this work, we propose an open-hardware low-power coarse-grained reconfigurable array connected to a lightweight microcontroller and enclosed in an application mapping framework. The latter provides complete support to configure kernels in the ...
- extended-abstractAugust 2023
X-HEEP: An Open-Source, Configurable and Extendible RISC-V Microcontroller
- Pasquale Davide Schiavone,
- Simone Machetti,
- Miguel Peón-Quirós,
- Jose Miranda,
- Benoît Denkinger,
- Thomas Christoph Müller,
- Ruben Rodríguez,
- Saverio Nasturzio,
- David Atienza Alonso
CF '23: Proceedings of the 20th ACM International Conference on Computing FrontiersMay 2023, Pages 379–380https://doi.org/10.1145/3587135.3591431X-HEEP (eXtendable Heterogeneous Energy-Efficient Platform) is an open-source1, configurable, and extensible single-core RISC-V microcontroller developed at the Embedded Systems Laboratory (ESL) of EPFL for edge-computing platforms. X-HEEP can be used ...
- research-articleAugust 2023
Graphene-Based Wireless Agile Interconnects for Massive Heterogeneous Multi-Chip Processors
- Sergi Abadal,
- Robert Guirado,
- Hamidreza Taghvaee,
- Akshay Jain,
- Elana Pereira de Santana,
- Peter Haring Bolívar,
- Mohamed Saeed,
- Renato Negra,
- Zhenxing Wang,
- Kun-Ta Wang,
- Max C. Lemme,
- Joshua Klein,
- Marina Zapater,
- Alexandre Levisse,
- David Atienza,
- Davide Rossi,
- Francesco Conti,
- Martino Dazzi,
- Geethan Karunaratne,
- Irem Boybat,
- Abu Sebastian
IEEE Wireless Communications (IEEEWIRCOM), Volume 30, Issue 4August 2023, Pages 162–169https://doi.org/10.1109/MWC.010.2100561The main design principles in computer architecture have recently shifted from a monolithic scaling-driven approach to the development of heterogeneous architectures that tightly co-integrate multiple specialized processor and memory chiplets. In such ...
- research-articleJuly 2023
ALPINE: Analog In-Memory Acceleration With Tight Processor Integration for Deep Learning
- Joshua Klein,
- Irem Boybat,
- Yasir Mahmood Qureshi,
- Martino Dazzi,
- Alexandre Levisse,
- Giovanni Ansaloni,
- Marina Zapater,
- Abu Sebastian,
- David Atienza
IEEE Transactions on Computers (ITCO), Volume 72, Issue 7July 2023, Pages 1985–1998https://doi.org/10.1109/TC.2022.3230285Analog in-memory computing (AIMC) cores offers significant performance and energy benefits for neural network inference with respect to digital logic (e.g., CPUs). AIMCs accelerate matrix-vector multiplications, which dominate these applications’ ...
- research-articleJanuary 2023
System-Level Exploration of In-Package Wireless Communication for Multi-Chiplet Platforms
- Rafael Medina,
- Joshua Kein,
- Giovanni Ansaloni,
- Marina Zapater,
- Sergi Abadal,
- Eduard Alarcón,
- David Atienza
ASPDAC '23: Proceedings of the 28th Asia and South Pacific Design Automation ConferenceJanuary 2023, Pages 561–566https://doi.org/10.1145/3566097.3567952Multi-Chiplet architectures are being increasingly adopted to support the design of very large systems in a single package, facilitating the integration of heterogeneous components and improving manufacturing yield. However, chiplet-based solutions have ...
- research-articleJanuary 2023
TiC-SAT: Tightly-Coupled Systolic Accelerator for Transformers
ASPDAC '23: Proceedings of the 28th Asia and South Pacific Design Automation ConferenceJanuary 2023, Pages 657–663https://doi.org/10.1145/3566097.3567867Transformer models have achieved impressive results in various AI scenarios, ranging from vision to natural language processing. However, their computational complexity and their vast number of parameters hinder their implementations on resource-...
- research-articleDecember 2022
HDTorch: Accelerating Hyperdimensional Computing with GP-GPUs for Design Space Exploration
ICCAD '22: Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided DesignOctober 2022, Article No.: 83, Pages 1–8https://doi.org/10.1145/3508352.3549475The HyperDimensional Computing (HDC) Machine Learning (ML) paradigm is highly interesting for applications involving continuous, semi-supervised learning for long-term monitoring. However, its accuracy is not yet on par with other ML approaches, ...
- research-articleNovember 2022
A Hardware/Software Co-Design Vision for Deep Learning at the Edge
- Flavio Ponzina,
- Simone Machetti,
- Marco Rios,
- Benoît Walter Denkinger,
- Alexandre Levisse,
- Giovanni Ansaloni,
- Miguel Peón-Quirós,
- David Atienza
IEEE Micro (IMIC), Volume 42, Issue 6Nov.-Dec. 2022, Pages 48–54https://doi.org/10.1109/MM.2022.3195617The growing popularity of edgeAI requires novel solutions to support the deployment of compute-intense algorithms in embedded devices. In this article, we advocate for a holistic approach, where application-level transformations are jointly conceived with ...
- articleSeptember 2022
PyBNesian: An extensible python package for Bayesian networks
Neurocomputing (NEUROC), Volume 504, Issue CSep 2022, Pages 204–209https://doi.org/10.1016/j.neucom.2022.06.112AbstractBayesian networks are probabilistic graphical models that are commonly used to represent the uncertainty in data. The PyBNesian package provides an implementation for many different types of Bayesian network models and some variants, ...
- research-articleAugust 2022
VWR2A: a very-wide-register reconfigurable-array architecture for low-power embedded devices
DAC '22: Proceedings of the 59th ACM/IEEE Design Automation ConferenceJuly 2022, Pages 895–900https://doi.org/10.1145/3489517.3530980Edge-computing requires high-performance energy-efficient embedded systems. Fixed-function or custom accelerators, such as FFT or FIR filter engines, are very efficient at implementing a particular functionality for a given set of constraints. However, ...