Model of the Weak Reset Process in HfO x Resistive Memory for Deep Learning Frameworks
Autor: | Etienne Nowak, Damien Querlioz, Marc Bocquet, Axel Laborieux, Atreya Majumdar, Elisa Vianello, Jean-Michel Portal, Tifenn Hirtzlin, Jacques-Olivier Klein |
---|---|
Přispěvatelé: | Institut des Matériaux, de Microélectronique et des Nanosciences de Provence (IM2NP), Aix Marseille Université (AMU)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS), Commissariat à l'énergie atomique et aux énergies alternatives - Laboratoire d'Electronique et de Technologie de l'Information (CEA-LETI), Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA), Centre de Nanosciences et de Nanotechnologies (C2N), Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), ANR-18-CE24-0009,NEURONIC,Réseau Neuronal Binaire à base d'architecture hybride de mémoires intégrant des fonctions de calcul (CMOS/RRAM) pour la fusion de capteurs(2018) |
Jazyk: | angličtina |
Rok vydání: | 2021 |
Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Computer science FOS: Physical sciences Applied Physics (physics.app-ph) 01 natural sciences Machine Learning (cs.LG) [SPI]Engineering Sciences [physics] 0103 physical sciences Electronic engineering Electrical and Electronic Engineering ComputingMilieux_MISCELLANEOUS 010302 applied physics Artificial neural network business.industry Deep learning Process (computing) Physics - Applied Physics Electronic Optical and Magnetic Materials Resistive random-access memory Artificial intelligence business Reset (computing) MNIST database Random access AND gate |
Zdroj: | IEEE Transactions on Electron Devices IEEE Transactions on Electron Devices, 2021, 68 (10), pp.4925-4932. ⟨10.1109/TED.2021.3108479⟩ IEEE Transactions on Electron Devices, Institute of Electrical and Electronics Engineers, 2021, 68 (10), pp.4925-4932. ⟨10.1109/TED.2021.3108479⟩ |
ISSN: | 0018-9383 |
DOI: | 10.1109/TED.2021.3108479⟩ |
Popis: | The implementation of current deep learning training algorithms is power-hungry, due to data transfer between memory and logic units. Oxide-based resistive random access memories (RRAMs) are outstanding candidates to implement in-memory computing, which is less power-intensive. Their weak RESET regime is particularly attractive for learning, as it allows tuning the resistance of the devices with remarkable endurance. However, the resistive change behavior in this regime suffers from many fluctuations and is particularly challenging to model, especially in a way compatible with tools used for simulating deep learning. In this work, we present a model of the weak RESET process in hafnium oxide RRAM and integrate this model within the PyTorch deep learning framework. Validated on experiments on a hybrid CMOS/RRAM technology, our model reproduces both the noisy progressive behavior and the device-to-device (D2D) variability. We use this tool to train binarized neural networks (BNNs) for the MNIST handwritten digit recognition task and the CIFAR-10 object classification task. We simulate our model with and without various aspects of device imperfections to understand their impact on the training process and identify that the D2D variability is the most detrimental aspect. The framework can be used in the same manner for other types of memories to identify the device imperfections that cause the most degradation, which can, in turn, be used to optimize the devices to reduce the impact of these imperfections. |
Databáze: | OpenAIRE |
Externí odkaz: |