Designing Neuromorphic 3D-Circuits using Memristors

Recently, I discovered a new interesting research area: neuromorphic circuits. Seems like the most interesting electronic circuit component used for designing neuromorphic circuits is the memristor, known as being the 4th electronic circuit component near the resistor, capacitor, and inductor. Its first concept was proposed theoretically in 1971 by Professor Leon Chua at the University of California, Berkeley, and physically realized for the first time only in 2008 by HP.

I remember knowing some ex-colleague of mine from the SRH Hochschule Heidelberg when I was doing an Erasmus semester there in 2013, which was doing some basic experiments with memristors for his Master thesis, but I didn’t take it too seriously at that time, because it was relatively new circuit component, and I was more busy with the software side (web design) that time. Despite some knowledge in AI developed during my Ph.D. and Post-Doc, I was always playing with the software side of AI but never implemented neuromorphic circuits, especially not ones with resistive devices (memristors). These days I was reading some research papers about how one can design some basic neuromorphic circuits using memristors and got a headache. I also installed LTspice, which seems like it is used for simulating simple neuromorphic circuit prototypes (also Cadence Virtuoso I saw being used often). I remember I had it installed in 2013 in Heidelberg, but never since due to me being a non-visionary at that time, haha.

A guy who does research in this field told me „For the circuits, I use LTSPice with compact models of memristors for prototyping. And then I use Cadence Analog mixed signals design toolchain to use the PDK and then transfer and fine-tune the transistor size etc.“ Also, „in order to stack them vertically, seems like if you want to have a working prototype, currently the only option is the Skywater 130nm process from Google„. Oh, Tesla, where are you?! I cry inside.

Below, I will post some videos and links I saved these days in case you are also interested in knowing more about this topic:

Neuromorphic 3D-Circuits – Google Suche – https://www.google.com/search?q=Neuromorphic+3D-Circuits&rlz=1C1CHBF_enDE1035DE1035&sourceid=chrome&ie=UTF-8#ip=1

Three-dimensional hybrid circuits: the future of neuromorphic computing hardware – https://iopscience.iop.org/article/10.1088/2632-959X/ac280e

(1) (PDF) Neuromorphic 3D Integrated Circuit: A Hybrid, Reliable and Energy Efficient Approach for Next Generation Computing – https://www.researchgate.net/publication/317051142_Neuromorphic_3D_Integrated_Circuit_A_Hybrid_Reliable_and_Energy_Efficient_Approach_for_Next_Generation_Computing

Monolithic 3D neuromorphic computing system with hybrid CMOS and memristor-based synapses and neurons – ScienceDirect – https://www.sciencedirect.com/science/article/abs/pii/S0167926017303413

2103.04852.pdf – https://arxiv.org/ftp/arxiv/papers/2103/2103.04852.pdf

Three-Dimensional Neuromorphic Computing System With Two-Layer and Low-Variation Memristive Synapses | IEEE Journals & Magazine | IEEE Xplore – https://ieeexplore.ieee.org/document/9360870

3D Neuromorphic Hardware with Single Thin‐Film Transistor Synapses Over Single Thin‐Body Transistor Neurons by Monolithic Vertical Integration – PMC – https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10602577/

Fully Integrated Memristor System for Neuromorphic and Analog Computing – https://apps.dtic.mil/sti/trecms/pdf/AD1171350.pdf

PowerPoint Presentation – https://www.mics.ece.vt.edu/content/dam/mics_ece_vt_edu/People/report/QE_Slide_HYA.pdf

TODAES2501-08 – https://www.cse.cuhk.edu.hk/~byu/papers/J42-TODAES2019-3D-FCN.pdf

3D-Integrated Neuromorphic Hardware With A Two-Level Neuromorphic “Synapse Over Neuron” Structure – https://semiengineering.com/3d-integrated-neuromorphic-hardware-with-a-two-level-neuromorphic-synapse-over-neuron-structure/

Brain-derived neuromorphic computing with 3D electronic-photonic integrated circuits| Yoo | Publications | Spie – https://spie.org/Publications/Proceedings/Paper/10.1117/12.2651109?SSO=1

3D neuromorphic circuits | Electrical and Electronic Engineering Community – https://engineeringcommunity.nature.com/posts/64163-3d-neuromorphic-circuits

Bio‐Inspired 3D Artificial Neuromorphic Circuits – Liu – 2022 – Advanced Functional Materials – Wiley Online Library – https://onlinelibrary.wiley.com/doi/abs/10.1002/adfm.202113050

Explore Neuromorphic Computing: Community, Education, and Research – https://open-neuromorphic.org/

sandialabs/cross-sim: CrossSim: accuracy simulation of analog in-memory computing – https://github.com/sandialabs/cross-sim

Microsoft Word – Panzeri_preprint_post.docx – https://arxiv.org/ftp/arxiv/papers/2212/2212.04317.pdf

Journal of Circuits, Systems and Computers – https://www.worldscientific.com/doi/pdf/10.1142/S0218126624501007?download=true

The Roadmap to Realize Memristive Three-Dimensional Neuromorphic Computing System | IntechOpen – https://www.intechopen.com/chapters/62198

Materials | Free Full-Text | Memristors for Neuromorphic Circuits and Artificial Intelligence Applications – https://www.mdpi.com/1996-1944/13/4/938

ICCAD 2023 – Power-Aware Training for Energy-Efficient Printed Neuromorphic CIrcuits – YouTube – https://www.youtube.com/watch?v=w-LW_8nwX9M

Energy Efficient AI Hardware: neuromorphic circuits and tools – YouTube – https://www.youtube.com/watch?v=PT0TO2pTrMo

Machine learning using magnetic stochastic synapses – IOPscience – https://iopscience.iop.org/article/10.1088/2634-4386/acdb96

memristor explained – YouTube – https://www.youtube.com/results?search_query=memristor+explained

Memristor devices ( presentation) – YouTube – https://www.youtube.com/watch?v=qAS0uCb6JxE

Neuromorphic Computing Explained | Jeffrey Shainline and Lex Fridman – YouTube – https://www.youtube.com/watch?v=u22-2CTErIQ

neuromorphic computing – YouTube – https://www.youtube.com/results?search_query=neuromorphic+computing

Brain-Like (Neuromorphic) Computing – Computerphile – YouTube – https://www.youtube.com/watch?v=Qow8pIvExH4

Bio-inspired Computing with Memristors – YouTube – https://www.youtube.com/watch?v=P2NO2lJyhkk

A review of memristor: material and structure design, device performance, applications and prospects – PMC – https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9980037/

Frontiers | Overview of Memristor-Based Neural Network Design and Applications – https://www.frontiersin.org/articles/10.3389/fphy.2022.839243/full

3D Convolutional Neural Network based on memristor for video recognition – ScienceDirect – https://www.sciencedirect.com/science/article/abs/pii/S0167865518309164

Day-14_Video-1 Neuromorphic Computing by Prof. Shubham Sahay – YouTube – https://www.youtube.com/watch?v=C2PYrPIPhQY

Flexible Simulation for Neuromorphic Circuit Design: Motion Detection Case Study – https://hal.science/hal-01538449/document

Neuromorphic Hardware – https://www.iis.fraunhofer.de/en/ff/kom/ai/neuromorphic.html

Handbook of Memristor Networks | SpringerLink – https://link.springer.com/book/10.1007/978-3-319-76375-0

J. Grollier – Neuromorphic computing: overview and challenges – YouTube – https://www.youtube.com/watch?v=yfnJqVHmJEw

Ocasys – https://ocasys.rug.nl/current/catalog/course/WMPH044-05

Mike Davies – New Tools for a New Era in Neuromorphic Computing – YouTube – https://www.youtube.com/watch?v=Fnf9yewGg1w

Joshua Yang: Memristive Materials and Devices for Neuromorphic Computing – YouTube – https://www.youtube.com/watch?v=4CI9Z0DM-AE

Achieving Green AI with Energy-Efficient Deep Learning Using Neuromorphic Computing | July 2023 | Communications of the ACM – https://cacm.acm.org/magazines/2023/7/274054-achieving-green-ai-with-energy-efficient-deep-learning-using-neuromorphic-computing/fulltext

Networking Neuromorphic Engineers – – https://www.neuropac.info/

Training Spiking Neural Networks Using Lessons From Deep Learning: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10242251

10 minutes paper (episode 4); Spiking NN – YouTube – https://www.youtube.com/watch?v=9dYZXQl4ozk

Cosyne 2022 Tutorial on Spiking Neural Networks – Part 1/2 – YouTube – https://www.youtube.com/watch?v=GTXTQ_sOxak

Neuromorphic computing with memristors: from device to system – Professor Huaqiang Wu – YouTube – https://www.youtube.com/watch?v=yChwSOXO538

Neuromorphic: BRAINLIKE Computers – YouTube – https://www.youtube.com/watch?v=19fjsk9blB4

ESWEEK 2021 Education – Spiking Neural Networks – YouTube – https://www.youtube.com/watch?v=7TybETlCslM

Prof. J. Joshua Yang, Memristive Devices for Neuromorphic Computing – YouTube – https://www.youtube.com/watch?v=35vu8qFy3Ys

Online dynamical learning and sequence memory with neuromorphic nanowire networks | Nature Communications – https://www.nature.com/articles/s41467-023-42470-5?

Neuromorphic learning, working memory, and metaplasticity in nanowire networks | Science Advances – https://www.science.org/doi/full/10.1126/sciadv.adg3289

tinyML EMEA 2021 Tutorial: Bio-inspired neuromorphic circuits architectures – YouTube – https://www.youtube.com/watch?v=aHHlBFqS99Y

Leben und Elektronik – Was kann ein Memristor? – YouTube – https://www.youtube.com/watch?v=sp9IZpAe8X0

Approaching the Area of Neuromorphic Computing Circuit and System Design | IEEE Conference Publication | IEEE Xplore – https://ieeexplore.ieee.org/document/9651627

Memristor Discovery (Board, Chip, Software, Manual) – Knowm Inc – https://knowm.com/collections/frontpage/products/memristor-discovery-board-chip-manual

DYNAP™-CNN: Neuromorphic Processor With 1M Spiking neurons – https://www.synsense.ai/products/dynap-cnn/

Old 2010 article about memristor – https://arxiv.org/ftp/arxiv/papers/1008/1008.2836.pdf

On-device machine learning with memristors in the neuromorphic era – YouTube – https://www.youtube.com/watch?v=bjcJg-bwBkU

Memristor-based Deep Spiking Neural Network with a Computing-In-Memory Architecture – YouTube – https://www.youtube.com/watch?v=hmsRDjk2lX0

A fully hardware-based memristive multilayer neural network | Science Advances – https://www.science.org/doi/10.1126/sciadv.abj4801

Making Memristive Neural Network Accelerators Reliable – https://www.bfeinberg.com/hpca18.pdf

Toolflow for the algorithm-hardware co-design of memristive ANN accelerators – ScienceDirect – https://www.sciencedirect.com/science/article/pii/S2773064623000439

[REFAI Seminar 09/30/21] Circuit Design & Silicon Prototypes for Compute-in-Memory for Deep Learning – YouTube – https://www.youtube.com/watch?v=vBvBiCZC2tw

Novel Memristor-based Neural Network Accelerators for Space Applications | Activities Portal – https://activities.esa.int/4000140774

Electronics | Free Full-Text | A Unified and Open LTSPICE Memristor Model Library – https://www.mdpi.com/2079-9292/10/13/1594

HOW NEUROMORPHIC COMPUTING WILL ACCELERATE ARTIFICIAL INTELLIGENCE – PROF SHUBHAM SAHAY- IIT KANPUR – YouTube – https://www.youtube.com/watch?v=sMjkG0jGCBs

Colloquium: Memristive Neuromorphic Computing Beyond Moore’s Law – YouTube – https://www.youtube.com/watch?v=DK_gzUujnXo

Neural Networks and Memristive Hardware Accelerators — Chair of Fundamentals of Electrical Engineering — TU Dresden – https://tu-dresden.de/ing/elektrotechnik/iee/ge/studium/lehrveranstaltungen/neural-networks-and-memristive-hardware-accelerators

Memristive devices based hardware for unlabeled data processing – IOPscience – https://iopscience.iop.org/article/10.1088/2634-4386/ac734a/meta

Memristor-based Spiking Neural
Networks.pdf – https://eprints.soton.ac.uk/471765/1/PhD_Thesis_1_.pdf

21 March 2023 Overview of Neuromorphic Computing Challenges and Opportunities by Sakib Hasan PhD – YouTube – https://www.youtube.com/watch?v=fSOpdgQXIao

Text Classification in Memristor-based Spiking
Neural Networks.pdf – https://arxiv.org/pdf/2207.13729.pdf

Recent progresses of in-memory computing: materials, devices and architectures – YouTube – https://www.youtube.com/watch?v=V3AQytpXI_s

Bill Dally – Trends in Deep Learning Hardware – YouTube – https://www.youtube.com/watch?v=4F_vfPFNe04

Training Spiking Neural Networks Using Lessons From Deep Learning – YouTube – https://www.youtube.com/watch?v=zldal7b7sJ4

Tutorial on snnTorch: Jason Eshraghian ICONS 2021 – YouTube – https://www.youtube.com/watch?v=O2-mT291ygg

Halide Perovskite Memristors for Neuromorphic Computing and Hardware Security – Rohit Abraham John – YouTube – https://www.youtube.com/watch?v=umaosrVFwk4

Roadmap on Neuromorphic Computing and Engineering – YouTube – https://www.youtube.com/watch?v=cR8I39rUWyM

The Basics of Neuromorphic Computing – YouTube – https://www.youtube.com/watch?v=o59pL5O4mXY

Spiking Neural Network – SNN – YouTube – https://www.youtube.com/watch?v=CJ_x9jYY4JU

Sensors | Free Full-Text | Overview of Spiking Neural Network Learning Approaches and Their Computational Complexities – https://www.mdpi.com/1424-8220/23/6/3037

Chinese Scientists Develop Groundbreaking AI Chip Inspired by the Brain – YouTube – https://www.youtube.com/watch?v=Ae6Eel7atXA

Edge learning using a fully integrated neuro-inspired memristor chip | Science – https://www.science.org/doi/10.1126/science.ade3483

Neuromorphic computing with emerging memory devices – YouTube – https://www.youtube.com/watch?v=gX9NqDuwTnA

‘Mind-blowing’ IBM chip speeds up AI – https://www.nature.com/articles/d41586-023-03267-0

Deep Learning Hardware – YouTube – https://www.youtube.com/watch?v=AGcv_PRKrPQ

tinyML Summit 2023: Tiny spiking AI for the sensor-edge – YouTube – https://www.youtube.com/watch?v=p9n6Pi46wT0

Low-Power Spiking Neural Network Processing Systems for Extreme-Edge Applications – Federico Corradi – YouTube – https://www.youtube.com/watch?v=xiYUVzdwDIA

In-Memory Computing based Machine Learning Accelerators: Opportunities and Challenges – YouTube – https://www.youtube.com/watch?v=TP2LVkgWCVE

Why spiking neural networks are important – Simon Thorpe, CERCO – YouTube – https://www.youtube.com/watch?v=8K5oc4y0Vas

Modeling and Demonstration of Hardware-based Deep Neural Network(DNN) using Memristor Crossbar Array – YouTube – https://www.youtube.com/watch?v=eEYf0110dtw

Leave a Comment

Diese Website verwendet Akismet, um Spam zu reduzieren. Erfahren Sie mehr darüber, wie Ihre Kommentardaten verarbeitet werden .