Matrix-Vector Multiplications (MVMs) represent a heavy workload for both training and inference in Deep Neural Networks (DNNs) applications. Analog In-memory Computing (AIMC) systems based on Phase Change Memory (PCM) has been shown to be a valid competitor to enhance the energy efficiency of DNN accelerators. Although DNNs are quite resilient to computation inaccuracies, PCM non-idealities could strongly affect MVM operations precision, and thus the accuracy of DNNs. In this paper, a combined hardware and software solution to mitigate the impact of PCM non-idealities is presented. The drift of PCM cells conductance is compensated at the circuit level through the introduction of a conductance ratio at the core of the MVM computation. A model of the behaviour of PCM cells is employed to develop a device-aware training for DNNs and the accuracy is estimated in a CIFAR-10 classification task. This work is supported by a PCM-based AIMC prototype, designed in a 90-nm STMicroelectronics technology, and conceived to perform Multiply-and-Accumulate (MAC) computations, which are the kernel of MVMs. Results show that the MAC computation accuracy is around 95% even under the effect of cells drift. The use of a device-aware DNN training makes the networks less sensitive to weight variability, with a 15% increase in classification accuracy over a conventionally-trained Lenet-5 DNN, and a 36% gain when drift compensation is applied.

Combined HW/SW Drift and Variability Mitigation for PCM-based Analog In-memory Computing for Neural Network Applications

Alessio Antolini
;
Francesco Zavalloni;Andrea Lico;Eleonora Franchi;Mauro Mangia;Riccardo Rovatti;
2023

Abstract

Matrix-Vector Multiplications (MVMs) represent a heavy workload for both training and inference in Deep Neural Networks (DNNs) applications. Analog In-memory Computing (AIMC) systems based on Phase Change Memory (PCM) has been shown to be a valid competitor to enhance the energy efficiency of DNN accelerators. Although DNNs are quite resilient to computation inaccuracies, PCM non-idealities could strongly affect MVM operations precision, and thus the accuracy of DNNs. In this paper, a combined hardware and software solution to mitigate the impact of PCM non-idealities is presented. The drift of PCM cells conductance is compensated at the circuit level through the introduction of a conductance ratio at the core of the MVM computation. A model of the behaviour of PCM cells is employed to develop a device-aware training for DNNs and the accuracy is estimated in a CIFAR-10 classification task. This work is supported by a PCM-based AIMC prototype, designed in a 90-nm STMicroelectronics technology, and conceived to perform Multiply-and-Accumulate (MAC) computations, which are the kernel of MVMs. Results show that the MAC computation accuracy is around 95% even under the effect of cells drift. The use of a device-aware DNN training makes the networks less sensitive to weight variability, with a 15% increase in classification accuracy over a conventionally-trained Lenet-5 DNN, and a 36% gain when drift compensation is applied.
2023
Alessio Antolini, Carmine Paolino, Francesco Zavalloni, Andrea Lico, Eleonora Franchi Scarselli, Mauro Mangia, Fabio Pareschi, Gianluca Setti, Riccardo Rovatti, Mattia Luigi Torres, Marcella Carissimi, Marco Pasotti
File in questo prodotto:
File Dimensione Formato  
Combined_HW_SW_Drift_and_Variability_Mitigation_for_PCM-based_Analog_In-memory_Computing_for_Neural_Network_Applications_IRIS.pdf

accesso aperto

Tipo: Versione (PDF) editoriale
Licenza: Creative commons
Dimensione 10.14 MB
Formato Adobe PDF
10.14 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/913711
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 6
  • ???jsp.display-item.citation.isi??? ND
social impact