Pasar al contenido principal

TECNOLOGIAS Y PLATAFORMAS EMERGENTES PARA COMPUTACION NEUROMORFICA EN ESPACIO - UPC

PID2022-141391OB-C22

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2022
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Programming Techniques of Resistive Random-Access Memory Devices for Neuromorphic Computing

  • Machado, Pau
  • Manich, Salvador
  • Gómez-Pau, Álvaro
  • Rodríguez-Montañés, Rosa
  • Bargalló González, Mireia
  • Campabadal, Francesca
  • Arumí, Daniel
Neuromorphic computing offers a promising solution to overcome the von Neumann bottleneck, where the separation between the memory and the processor poses increasing limitations of latency and power consumption. For this purpose, a device with analog switching for weight update is necessary to implement neuromorphic applications. In the diversity of emerging devices postulated as synaptic elements in neural networks, RRAM emerges as a standout candidate for its ability to tune its resistance. The learning accuracy of a neural network is directly related to the linearity and symmetry of the weight update behavior of the synaptic element. However, it is challenging to obtain such a linear and symmetrical behavior with RRAM devices. Thus, extensive research is currently devoted at different levels, from material to device engineering, to improve the linearity and symmetry of the conductance update of RRAM devices. In this work, the experimental results based on different programming pulse conditions of RRAM devices are presented, considering both voltage and current pulses. Their suitability for application as analog RRAM-based synaptic devices for neuromorphic computing is analyzed by computing an asymmetric nonlinearity factor., This work was supported in part by Spanish MCIN/AEI /10.13039/501100011033/FEDER, under Projects PID2022-141391OB-C22 and PID2022-139586NB-C42., Peer reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

A spatio-temporal-based concept for associative memory modeling with memristors

  • Chatzipaschalis, Ioannis|||0000-0002-3468-4442
  • Tompris, Ioannis
  • Stavroulakis, Emmanouil
  • Chatzinikolaou, Theodoros Panagiotis
  • Fyrigos, Iosif-Angelos
  • Fraidakis, Pantelis
  • Calomarde Palomino, Antonio|||0000-0002-4459-8505
  • Gomá Torrellas, Rafael
  • Sirakoulis, Georgios Ch.
  • Rubio Sola, Jose Antonio|||0000-0003-1625-1472
Spatio-temporal encoding in neural systems refers to the representation of information through the combined spatial and temporal patterns of neuronal activity and plays a fundamental role in various brain biological functions, such as higher cognitive processes, including the memory effect. Memory is an intricate process in human cognition and this bio-inspired function can be replicated by novel hardware components. Memristors can imitate the memory effect as they appear to have non-volatile characteristics, and they can also constitute the switching component of memristive-based oscillators acting as neurons for designing complex network structures in order to replicate the human brain behavior. In this paper, a spatio-temporal encoding scheme is being evaluated under a hardware-oriented concept integrating memristive novel devices, aiming to replicate the way associative memory works, and paving the way for novel approaches in brain-inspired computing systems., This work was supported in part by the Spanish Ministry of Science, Innovation and Universities (MICIU) under Grant PID2022-141391OBC22/AEI/10.13039/501100011033, and in part by FI SDUR co-financed by AGAUR and the European Union through the European Social Fund Plus (ESF+)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Character recognition application of a neural circuit including lateral inhibitory mechanisms

  • Llobet, Dídac
  • Chatzipaschalis, Ioannis|||0000-0002-3468-4442
  • Calomarde Palomino, Antonio|||0000-0002-4459-8505
  • Rubio Sola, Jose Antonio|||0000-0003-1625-1472
This article presents the hardware simulation of neurons using memristive devices as synapses. To model the neuronal behavior of a nerve, that is, a coherent set of neurons, we introduce two well-known mechanisms from biology: lateral inhibition and threshold adaptation to repetitive excitations. We present circuit simulations and a high-level model that demonstrate scalable behavior. This network demonstrates the ability to learn and recognize characters, using lateral inhibition as a key dynamizing element., This work was supported in part by Spanish MCIN/AEI/10.13039/501100011033/FEDER, under Project PID2022- 141391OB-C22 and in part by Spanish Ministry of “Transformacion Digital y Funcion Publica”, NextGenerationEU, TSI-069100-2023-0015., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

A spatio-temporal-based concept for associative memory modeling with memristors

  • Chatzipaschalis, Ioannis
  • Tompris, Ioannis
  • Stavroulakis, Emmanouil
  • Chatzinikolaou, Theodoros Panagiotis
  • Fyrigos, Iosif-Angelos
  • Fraidakis, Pantelis
  • Calomarde Palomino, Antonio
  • Gomá Torrellas, Rafael
  • Sirakoulis, Georgios Ch.
  • Rubio Sola, Jose Antonio
Spatio-temporal encoding in neural systems refers to the representation of information through the combined spatial and temporal patterns of neuronal activity and plays a fundamental role in various brain biological functions, such as higher cognitive processes, including the memory effect. Memory is an intricate process in human cognition and this bio-inspired function can be replicated by novel hardware components. Memristors can imitate the memory effect as they appear to have non-volatile characteristics, and they can also constitute the switching component of memristive-based oscillators acting as neurons for designing complex network structures in order to replicate the human brain behavior. In this paper, a spatio-temporal encoding scheme is being evaluated under a hardware-oriented concept integrating memristive novel devices, aiming to replicate the way associative memory works, and paving the way for novel approaches in brain-inspired computing systems., This work was supported in part by the Spanish Ministry of Science, Innovation and Universities (MICIU) under Grant PID2022-141391OBC22/AEI/10.13039/501100011033, and in part by FI SDUR co-financed by AGAUR and the European Union through the European Social Fund Plus (ESF+)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Programming techniques of resistive random-access memory devices for neuromorphic computing

  • Machado Panadés, Pau
  • Manich Bou, Salvador
  • Gómez Pau, Álvaro
  • Rodríguez Montañés, Rosa
  • Bargalló González, Mireia
  • Campabadal, Francesca
  • Arumi Delgado, Daniel
Neuromorphic computing offers a promising solution to overcome the von Neumann bottleneck, where the separation between the memory and the processor poses increasing limitations of latency and power consumption. For this purpose, a device with analog switching for weight update is necessary to implement neuromorphic applications. In the diversity of emerging devices postulated as synaptic elements in neural networks, RRAM emerges as a standout candidate for its ability to tune its resistance. The learning accuracy of a neural network is directly related to the linearity and symmetry of the weight update behavior of the synaptic element. However, it is challenging to obtain such a linear and symmetrical behavior with RRAM devices. Thus, extensive research is currently devoted at different levels, from material to device engineering, to improve the linearity and symmetry of the conductance update of RRAM devices. In this work, the experimental results based on different programming pulse conditions of RRAM devices are presented, considering both voltage and current pulses. Their suitability for application as analog RRAM-based synaptic devices for neuromorphic computing is analyzed by computing an asymmetric nonlinearity factor., This work was supported in part by Spanish MCIN/AEI /10.13039/501100011033/FEDER, under Projects PID2022-141391OB-C22 and PID2022-139586NB-C42., Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Character recognition application of a neural circuit including lateral inhibitory mechanisms

  • Llobet, Dídac
  • Chatzipaschalis, Ioannis
  • Calomarde Palomino, Antonio
  • Rubio Sola, Jose Antonio
This article presents the hardware simulation of neurons using memristive devices as synapses. To model the neuronal behavior of a nerve, that is, a coherent set of neurons, we introduce two well-known mechanisms from biology: lateral inhibition and threshold adaptation to repetitive excitations. We present circuit simulations and a high-level model that demonstrate scalable behavior. This network demonstrates the ability to learn and recognize characters, using lateral inhibition as a key dynamizing element., This work was supported in part by Spanish MCIN/AEI/10.13039/501100011033/FEDER, under Project PID2022- 141391OB-C22 and in part by Spanish Ministry of “Transformacion Digital y Funcion Publica”, NextGenerationEU, TSI-069100-2023-0015., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Memristor-based cellular automata for natural language processing

  • Chatzipaschalis, Ioannis
  • Chatzinikolaou, Theodoros Panagiotis
  • Fyrigos, Iosif-Angelos
  • Adamatzky, Andrew
  • Rubio Sola, Jose Antonio
  • Sirakoulis, Georgios Ch.
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Towards achieving efficient computational approaches, the perspective of Cellular Automata (CAs) appears to be highly potent, owing to its performance hidden in the parallel computing capabilities inherent in its local units, known as cells. Through their local interconnections and adherence to fixed or dynamic rules, these cells demonstrate exceptional capability in executing intricate physical events and addressing emergent problems that pose significant challenges for conventional computing systems. Capitalizing on the non-volatility and information processing capabilities offered by memristors, an oscillatory memristive CA circuit is proposed aiming to advance the field of neuromorphic computing and artificial intelligence applications, as well as to point out more effective and reliable computational models by bridging the gap between biological and circuit-based networks. In this manner, CAs are introduced as nanoelectronic memristor-based circuits to become a good candidate for utilizing Natural Language Processing (NLP) tasks like the forming of sentences., This work has been partially supported by the framework of the FUNGATERIA project, which has received funding from the European Union’s HORIZON-EIC-2021-PATHFINDER CHALLENGES programme under grant agreement No. 101071145, and by the project PID2022- 141391OB-C22, Ministry of Science, Spain., Peer Reviewed, Postprint (published version)