In-Memory Computing for Machine Learning and Deep Learning

Lepri, Nicola; Glukhov, Artem; Cattaneo, Lorenzo; Farronato, Matteo; Mannocci, Piergiulio; Ielmini, Daniele

doi:10.1109/jeds.2023.3265875

Cited by 5 publications

(2 citation statements)

References 175 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…of Electrical Engineering and Information Systems, The University of Tokyo, 113-8656, Japan. a) E-mail: ichikawa@co-design.t.u-tokyo.ac.jp overhead of peripheral circuits such as DAC/ADC [17]. The capacities of recently reported CiMs are around 1-10 Mb [4,[18][19][20].…”

Section: Introductionmentioning

confidence: 99%

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

ICHIKAWA,

YAMADA,

MISAWA

et al. 2024

IEICE Trans. Electron.

View full text Add to dashboard Cite

Integrating RGB and event sensors improves object detection accuracy, especially during the night, due to the high-dynamic range of event camera. However, introducing an event sensor leads to an increase in computational resources, which makes the implementation of RGB-event fusion multi-modal AI to CiM difficult. To tackle this issue, this paper proposes RGB-Event fusion Multi-modal analog Computation-in-Memory (CiM), called REM-CiM, for multi-modal edge object detection AI. In REM-CiM, two proposals about multi-modal AI algorithms and circuit implementation are co-designed. First, Memory capacity-Efficient Attentional Feature Pyramid Network (MEA-FPN), the model architecture for RGB-event fusion analog CiM, is proposed for parameter-efficient RGB-event fusion. Convolution-less bi-directional calibration (C-BDC) in MEA-FPN. extract important features of each modality with attention modules, while reducing the number of weight parameters by removing large convolutional operations from conventional BDC. Proposed MEA-FPN w/ C-BDC achieves a 76% reduction of parameters while maintaining mean Average Precision (mAP) degradation to <2.3% during both day and night, compared with Attentional FPN fusion (A-FPN), a conventional BDC-adopted FPN fusion. Second, the low-bit quantization with clipping (LQC) is proposed to reduce area/energy. Proposed REM-CiM with MEA-FPN and LQC achieves almost the same memory cells, 21% less ADC area, 24% less ADC energy and 0.7% higher mAP than conventional FPN fusion CiM without LQC.

show abstract

Section: Introductionmentioning

confidence: 99%

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

ICHIKAWA,

YAMADA,

MISAWA

et al. 2024

IEICE Trans. Electron.

View full text Add to dashboard Cite

show abstract

“…Neuromorphic computing systems that emulate the biophysics of neurons and synapses in the human brain are expected to have very high energy efficiency and might be thus potential candidates for the new generation of beyond von Neumann paradigms [3,4]. The human brain consists of a great number of neurons and synapses, which are interconnected.…”

Section: Introductionmentioning

confidence: 99%

Analog memristive devices based on La₂NiO₄₊ _δ as synapses for spiking neural networks

Khuu,

Koroleva,

Degreze

et al. 2023

J. Phys. D: Appl. Phys.

View full text Add to dashboard Cite

Neuromorphic computing has recently emerged as a potential alternative to the conventional von Neumann computer paradigm, which is inherently limited due to its architectural bottleneck. Thus, new artificial components and architectures for brain-inspired computing hardware implementation are required. Bipolar analog memristive devices, whose resistance (or conductance) can be continuously tuned (as a synaptic weight), are potential candidates for artificial synapse applications. In this work, lanthanum nickelate (La2NiO4+δ , L2NO4), a mixed ionic electronic conducting oxide, is used in combination with TiN and Pt electrodes. The TiN/L2NO4/Pt devices show bipolar resistive switching with gradual transitions both for the SET and RESET processes. The resistance (conductance) can be gradually modulated by the pulse amplitude and duration, showing good data retention characteristics. A linear relationship between the resistance change and total applied pulse duration is experimentally measured. Moreover, synaptic depression and potentiation characteristics, one of the important functions of bio-synapses, are artificially reproduced for these devices, then modeled and successfully tested in a spiking neural network environment. These results indicate the suitability of using TiN/L2NO4/Pt memristive devices as long-term artificial synapses in neuromorphic computing.

show abstract

Implementation of Boolean Logic Functions Through Double Gate FD-SOI MOSFET

Ansari,

El-Atab

2024

IEEE Access

View full text Add to dashboard Cite

In-Memory Computing for Machine Learning and Deep Learning

Cited by 5 publications

References 175 publications

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

Analog memristive devices based on La₂NiO₄₊ _δ as synapses for spiking neural networks

Implementation of Boolean Logic Functions Through Double Gate FD-SOI MOSFET

Contact Info

Product

Resources

About

In-Memory Computing for Machine Learning and Deep Learning

Cited by 5 publications

References 175 publications

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

REM-CiM: Attentional RGB-Event Fusion Multi-Modal Analog CiM for Area/Energy-Efficient Edge Object Detection during Both Day and Night

Analog memristive devices based on La2NiO4+ δ as synapses for spiking neural networks

Implementation of Boolean Logic Functions Through Double Gate FD-SOI MOSFET

Contact Info

Product

Resources

About

Analog memristive devices based on La₂NiO₄₊ _δ as synapses for spiking neural networks