Neuromorphic Computing and Sensing 2021 - Sample
0 likes1,610 views
The report examines the neuromorphic computing and sensing market, detailing market forecasts and the technology landscape from 2020 to 2035. It highlights key applications in sectors such as mobile, automotive, medical, and industrial, and outlines trends in neuromorphic technologies, including processing and sensing. Additionally, it covers the ecosystem of major players and innovations within the field, emphasizing the integration of neuromorphic concepts in AI and computational frameworks.
Neuromorphic Computing and Sensing 2021 - Sample
- 1. From Technologies to Markets © 2021 From Technologies to Markets Neuromorphic Computing and Sensing Market and Technology Report 2021 Sample
- 2. 2 TABLE OF CONTENTS - 1/2 o Report objectives and scope 7 o Methodology 10 o 3 slides summary 16 o Executive summary 20 1-Context 48 o Moore law, more than Moore and data explosion o Limitation of the current paradigm o Definition o Main neuromorphic approaches 2-Market forecast 62 o Neuromorphic sensing and computing applications o Neuromorphic sensing and computing potential benefits o 2020 - 2035 Neuromorphic sensing and computing in $. o 2020 - 2035 Neuromorphic sensing and computing, split by market in %. o 2020 - 2035 System-level total available market in Munits. o 2020 - 2035 Neuromorphic technology penetration in %. o 2020 - 2035 Neuromorphic sensor average selling price in $. o 2020 - 2035 Neuromorphic processor average selling price in $. o What is new compared to the 2019 report? 3-Market trends 80 o Mobile and consumer o Automotive o Medical o Industrial o Datacenter 4-Ecosystem 163 o AI company landscape o Neuromorphic and in memory computing players – technology segmentation o Neuromorphic image sensor player mapping o Neuromorphic processor player mapping o Neuromorphic company market positioning Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 3. 3 TABLE OF CONTENTS - 2/2 5-Neuromorphic technologies 173 o Neuromorphic sensor technology • Image sensor technology trends • Event-based neuromorphic image sensors existing products o Neuromorphic computing technology • Neuromorphic computing processors key concepts • Neuromorphic software • Spiking Neural Networks • Software readiness and acceptance • Available processors • Top 20 patent-holding companies 2019 • Player's description o Emerging non-volatile memories for neuromorphic computing • Neuromorphic in-memory computing concept • Emerging NVM, ecosystem and roadmap • Forecast for emerging NVM • Player's description • Embedded NVM for In-Memory Computing o Appendix 249 • AI definition • Processor definition • Yole Développement Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 4. 4 4 SCOPE OFTHE REPORT – ADOPTION AREAS OF NEUROMORPHIC Market Telecom & infrastructure Industrial Medical Automotive Mobile and consumer End system Machine vision for factory automation Robotics Surveillance camera Prosthetics Forward camera, In cabin imaging Laptop PCs,Tablets Personal assistant, personal robotics, consumer security camera Application Smart city Medical ADAS Client PC Phone Logistic and others Server Smartphone Wearables Home Smartwatches, earbuds Drone Consumer drones Manufacturing Datacenter Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 Robotic Vehicle Robotic vehicle cameras Drone Industrial drones Yours needs are out of the report’ scope? Contact us for a custom study: In this report we analyzed 5 end-markets, 13 applications and a variety of related end- systems
- 5. 5 5 SCOPE OFTHE REPORT – PLAYERS & SYSTEM APPROACHES The report focuses on neuromorphic IC chips for an efficient execution of AI algorithms AI chips The goal is to create processor which allow to run AI algorithm efficiently Neuroscience-focused/Brain simulator The goal is to create processors which allow to simulate the brain activity and to validate neuroscientific models. These elements can be utilized to design the architecture of a future neuromorphic AI accelerators. Neuromorphic computing Processors which are inspired from the brain Research driven project Go to market approach Main focus of the report Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 Non-exhaustive list And many others… Yours needs are out of the report’ scope? Contact us for a custom study: Courtesy of Intel
- 6. 6 6 • Definition and description of neuromorphic technologies. • 2020 - 2035 Neuromorphic sensing and computing market forecast in $. • 2020 - 2035 Neuromorphic technology penetration in %. • Mobile, automotive, medical, industrial and data center market trends. • Neuromorphic players – mapping and technology segmentation. • Available neuromorphic processors and image sensors. • Neuromorphic software description, software readiness, and acceptance. • Neuromorphic player's descriptions. • Neuromorphic in-memory computing concept. • Emerging non-volatile memory, ecosystem, and roadmap. • Forecast for emerging non-volatile memory. • Embedded non-volatile memory for in-memory computing. Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 KEY FEATURES OF REPORT
- 7. 7 METHODOLOGY AND DEFINITIONS Market segmentation volume in Munits, ASP in $, revenue in $M Yole’s market forecast model is based on matching heterogeneous sources: Information aggregation Preexisting information Forecast Reconciliation END SYSTEM SEMICONDUCTOR DEVICE Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 8. 8 8 Adrien Sanchez As a Technology & Market Analyst, Adrien Sanchez belongs to the Computing & Software business unit at Yole Développement (Yole), part of the Yole Group of Companies. In collaboration with his team, Adrien produces technology & market analyses covering computing hardware and software, AI, machine learning and neural networks. Prior to Yole, he worked as an intern at AW Europe (Belgium), where he focused on image recognition & comprehension for ADAS. He also worked at ACOEM (France), where he focused on real-time sound classification using deep learning and edge computing. Adrien graduated with a double degree at Grenoble Institute of Technology PHELMA (Grenoble INP Phelma, France) and Grenoble Ecole de Management (GEM, France), and he earned an MSc on AI at Heriot-Watt University (Edinburgh, UK). Contact: adrien.sanchez@yole.fr Pierre Cambou Pierre Cambou MSc, MBA, is a Principal Analyst in the Photonics and Sensing Division atYole Développement (Yole). Pierre’s mission is dedicated to imaging related activities by providing market & technology analyses along with strategy consulting services to semiconductor companies. He is responsible for the CIS Quarterly Market Monitor while he has authored more than 15 Yole Market & Technology reports. Pierre has an Engineering degree from Université de Technologie de Compiègne (France) and a Master of Science from Virginia Tech. (VA, USA), Pierre also graduated with an MBA from Grenoble Ecole de Management (France). Contact: pierre.cambou@yole.fr Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 Biographies & contact ABOUT THE AUTHORS Simone Bertolazzi,Ph.D. Simone is a Sr. Technology & Market analyst at Yole Développement (Yole) working with the Semiconductor & Software division. He is member of Yole’s memory team and contributes on a day-to-day basis to the analysis of memory markets and technologies, their related materials and fabrication processes. Previously, Simone carried out experimental research in the field of nanoscience and nanotechnology, focusing on emerging semiconducting materials and their device applications. He (co-) authored more than 15 papers in high-impact scientific journals and was awarded the prestigious Marie Curie Intra-European Fellowship. Simone obtained a PhD in physics in 2015 from École Polytechnique Fédérale de Lausanne (Switzerland), where he developed novel flash memory cells based on heterostructures of 2D materials and high-κ dielectrics. Simone earned a double M.A. Sc. degree from Polytechnique de Montréal (Canada) and Politecnico di Milano (Italy), graduating cum laude. Contact: simone.bertolazzi@yole.fr
- 9. 9 9 ABR,Accenture,Adesto Technologies, ai Storm,Allibaba.com,Amazon,Ambarella,AMD,AMT, Anotherbrain,Antaïos,Apple,Applied Materials,ARM,AryballeTechnologies,AvalancheTechnology, Aspinity,AWS, Axis,Azure, Baidu, BMW, Brainchip, Canon, Celepixel, Ceva, Crossbar, Cogito Instruments, Dahua, Dialog, Facebook, Fraunhofer, GeneralVision, GlobalFoundries, Google, Gorilla, GrAI Matter Lab, Groq, HPLabs, GyrfalconTechnology, HLMC, Hprobe, Huawei, IBM, IMEC, Infineon, IniLabs, Inivation, Innatera, Insightness, Intel, Kalray, Knowm, Mediatek, Microchip, Micron, Mythic, Natural Intelligence, Nepes AI, NeuroMem, Numem, Numenta, Nuvoton, Nvidia, Omnivision, Prophesee, Qualcomm, Rain Neuromorphics, Robosensing, Samsung, Scheider Electric, Sensigent, SMIC, Sony, ST Microelectronics, Synsense, Synthara, Syntiant,TDK,TSMC, UMC,Vicarious,Weebit,Westwell Labs, Xilinx and more… Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 COMPANIES CITED IN THIS REPORT
- 10. 10 10 2021 – 2035 – APPLICATION’S ADOPTION OF NEUROMORPHIC TECHNOLOGIES 2025 2030 Smartphone Industrial drones Smart home Wearables In-cabin cameras Prosthetics Security Industrial cameras MachineVision for robotics Consumer drones ADAS Server Industrial Consumer Automotive Medical Datacenter Volume shipments 2021 2035 Robotic vehicles From 2025 to 2030: Neuromorphic applications takes off Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 11. 11 11 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 FORECAST 2021-2035
- 12. 12 12 NEUROMORPHIC INTO AI COMPUTING & SENSING – 2025 – 2030 - 2035 REVENUE EVOLUTION 99,7% 0,3% 92% 8% 82% 18% 2035 2030 2025 Neuromorphic AI Non-neuromorphic AI We expect neuromorphic computing and sensing will represent between 15% and 20% of total AI computing revenue in 2035 ~$20B ~$7B ~$0.2B *Artificial Intelligence Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 13. 13 13 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 MARKET TRENDS – CONSUMER & AUTOMOTIVE
- 14. 14 14 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 MARKET TRENDS – MEDICAL & DATACENTER
- 15. 15 15 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 MARKET TRENDS - INDUSTRIAL
- 16. 16 16 ECOSYSTEM Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 17. 17 17 2021 NEUROMORPHIC SENSING AND COMPUTING ECOSYSTEM Universities Start-ups Semiconductor large companies Sensing Computing Non-exhaustive list of companies Sensing Computing Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 18. 18 18 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 NEUROMORPHIC COMPUTING PROCESSORS KEY CONCEPTS Parallel architecture Cores are distributed and operate in parallel. Memory, computation and communication integration All of these components are integrated together, as it is the case in the brain. Large scalability Plasticity Connection sparsity Activation sparsity For a specific decision, only a small portion of neurons are firing. It is neurons which are stimulated by inputs. Cores are communicating via an on-chip event-driven network. It is an event-driven behavior at the neuron/core level. A neuron is connected to only few neurons and few connection are responsible for the core of the computation (only for NN based approaches). It is a key element for scalability. Ability to learn online One neurosynaptic core represent an artificial neural. Chips communicate via an inter-chip interface leading to seamless availability like the brain’s cortex, enabling the creation of scalable neuromorphic systems. Neurosynaptic core Neuromorphic key concepts Hardware related concept Software related concept Software representation of neurons Clock-less processors Algorithms based on neural networks Event-driven Time dependency, make computation only when an event occur. Digital / Analog Learning Adaptability and fault tolerance Time-series processing This is a summary of these key neuromorphic concepts for a neuromorphic chip. It doesn’t mean that each neuromorphic processor implement all these concepts, as neuromorphic processors can mimic the brain according to several ways. But these concept are often at the hearth of innovations which are making chip neuromorphic. SNN-based processors are maid following most of these properties and will receive a specific attention.These concepts can be integrated into processors at different degree, it is not always binary properties. In all cases, a tight integration with software is critical for reaching power efficiency, and this is particularly true for neuromorphic computing. As time is an inherent component of spiking approaches, it open the door for efficient time series approach (both low power and high speed and accuracy)
- 19. 19 19 • The main goal of AI is for an artificial system to be autonomous. For that, understanding the environment and navigating through it makes up a large part of the research. • Machine Learning (ML) is one of the most important fields of AI. The key idea behind ML is teaching an algorithm how to solve a problem by showing it examples of the solution solved, without formally explaining how to solve the solution. Three main types of learning exist supervised learning, unsupervised learning and reinforcement learning. • An (artificial) neural network is a network of simple elements called neurons and is part of machine learning methods. Neurons are organized in layers, and when a network has several layers, it is called a Deep Neural Network (DNN). We are talking about Deep Learning (DL) when Machine Learning principles are applied to DNN. • Spiking Neural Network are a specific type of neural network which is using temporal spikes to encode the information. It can be used through ML principles for some applications such as classification. • Spiking Neural Network can also be used for non-Machine Learning application such as SLAM or stochastic CSP. Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 SPIKING NEURAL NETWORK IS A NEUROMORPHIC SOFTWARE Artificial Intelligence Machine Learning Neural Network SupportVector Machine (SVM) Decision Trees Nearest Neighbor Spiking Neural Network … Spiking Neural Network • SLAM • Evolutionary search • Stochastic CSP • … Software representation of neurons Spiking Neural Network are not the only software model compatible with neuromorphic computing, but it is the most spread paradigm in this field. 2nd generation deep Neural Network
- 20. 20 20 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 TECHNOLOGY TRENDS – SENSING & COMPUTING
- 21. 21 21 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 TECHNOLOGY TRENDS – NONVOLATILE MEMORY
- 22. 22 Contact our SalesTeam for more information 22 Contact our SalesTeam for more information MachineVision for Industry and Automation 2021 Emerging Non-Volatile Memory 2021 Cameras and Computing for Surveillance and Security 2020 Artificial Intelligence Computing for Automotive 2020 Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021 YOLE GROUP OF COMPANIES RELATED ANALYSES Yole Intelligence
- 23. 23 23 Yole Group of Companies, including Yole Développement, System Plus Consulting and PISEO, are pleased to provide you a glimpse of our accumulated knowledge. We invite you to share our data with your own network, within your presentations, press releases, dedicated articles and more, but you first need approval from Yole Public Relations department. If you are interested, feel free to contact us right now! We will also be more than happy to give you updated data and appropriate formats. Your contact: Sandrine Leroy, Dir. Public Relations Email: leroy@yole.fr HOWTO USE OUR DATA? Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021
- 24. 24 About Yole Développement | www.yole.fr | ©2020 CONTACTS Western US & Canada Steve Laferriere - steve.laferriere@yole.fr + 1 310 600 8267 Eastern US & Canada ChrisYouman - chris.youman@yole.fr +1 919 607 9839 Europe and RoW Lizzie Levenez - lizzie.levenez@yole.fr +49 15 123 544 182 Benelux, UK & Spain Marine Wybranietz - marine.wybranietz@yole.fr +49 69 96 21 76 78 India and RoA Takashi Onozawa - takashi.onozawa@yole.fr +81 80 4371 4887 Greater China MavisWang - mavis.wang@yole.fr +886 979 336 809 +86 136 6156 6824 Korea Peter Ok - peter.ok@yole.fr +82 10 4089 0233 Japan Miho Ohtake - miho.ohtake@yole.fr +81 34 4059 204 Japan and Singapore Itsuyo Oshiba - itsuyo.oshiba@yole.fr +81 80 3577 3042 Japan Toru Hosaka – toru.hosaka@yole.fr +81 90 1775 3866 FINANCIAL SERVICES › Jean-Christophe Eloy - eloy@yole.fr +33 4 72 83 01 80 › Ivan Donaldson - ivan.donaldson@yole.fr +1 208 850 3914 CUSTOM PROJECT SERVICES › Jérome Azémar, Yole Développement - jerome.azemar@yole.fr - +33 6 27 68 69 33 › Julie Coulon, System Plus Consulting - jcoulon@systemplus.fr - +33 2 72 17 89 85 GENERAL › Sandrine Leroy, Public Relations sandrine.leroy@yole.fr - +33 4 72 83 01 89 › General inquiries: info@yole.fr - +33 4 72 83 01 80 Follow us on REPORTS, MONITORS &TRACKS Neuromorphic Computing and Sensing 2021 | Sample | www.yole.fr | ©2021