Hardware Podcasts

Carver Mead Says Neuromorphic Efficiency Can Help AI

EE Times Current · 4 Oct 2024 · 1h 05m

What this episode covers

Caltech’s Professor Carver Mead, a founder of neuromorphic engineering and early Intel employee, discusses the origins of parallel processing, the transition from analog to digital AI, and his work at startups like Fauvian and Synaptics. He explains how his early work on scaling transistors and silicon compilers laid the groundwork for modern parallel computation, while his collaboration with Misha Mahowald led to the first event-based retinas. The episode also features commentary from Dr. Giulia D’Angelo and Prof. Ralph Etienne-Cummings on the current state of neuromorphic hardware, the inefficiencies of digital backpropagation, and the potential of dendritic computing for autonomous vehicles.

Most relevant to

established companies

Topics

Mentioned

Companies
Intel, Fauvian, Synaptics, Sigma, National Semiconductor, Caltech, University College London, Johns Hopkins University, Czech Technical University, Fairchild, NVIDIA, Carnegie Mellon, Harvard
Products
X3
Also
Misha Mahowald Prize

Related episodes from other shows

Other shows on the same ground, closest first.

  • The Circuit · 42m

    Episode 67: A Conversation with Aart de Geus on EDA and the Future of Silicon Design

    Aart de Geus, co-founder and executive chair of Synopsys, discusses the evolution of electronic design automation (EDA) and the future of silicon design. He explains how the industry shifted from analog to digital design, enabling the automation that drove Moore's Law, and details the transition to the Angstrom era with technologies like FinFETs and backside power delivery. De Geus also highlights the rise of systemic complexity through chiplets and vertical stacking, and the critical role of AI in optimizing these complex multi-die architectures.

    EDA & CAD toolsAI & machine learning
    8 chapters

  • IoT For All Podcast · 19m

    The Evolution of AI in IoT | BrainChip's Nandan Nayampally | Internet of Things Podcast

    Nandan Nayampally, CMO at BrainChip, discusses the evolution of AIoT and the shift of AI processing from the cloud to the edge. He explains how neuromorphic computing enables real-time, private, and efficient inference on billions of devices, citing PwC's $15 trillion GDP impact estimate. Nayampally highlights key adoption markets like healthcare, agriculture, and infrastructure, and advises companies to focus on solving acute pain points ('headaches') rather than nice-to-have features ('vitamins') to gain traction.

    AI & machine learningConnected products & IoT
    5 chapters

  • Go/No-Go · 54m

    Federico Faggin designed the first microprocessor and the capacitive touchscreen. He joins Go/No-Go to argue that AI cannot be conscious, however capable it becomes.

    Federico Faggin, inventor of the Intel 4004 microprocessor and the capacitive touchscreen, recounts his journey from developing MOS silicon gate technology at Fairchild Semiconductor to founding Zilog and Synaptics. He details the physical realities of early chip design, such as hand-drawing transistors and using slide rules, and explains how he built the Z80 with only $400,000 and 11 people. Faggin argues in his book Irreducible that classical computers and AI cannot be conscious because they lack the quantum field interactions necessary for true feeling and free will.

    Chips, FPGAs & semiconductorsAI & machine learning
    7 chapters

  • Embedded.fm · 59m

    376: Left Half of My Brain Is Digital

    David Comer, a retired engineer with a career spanning from the Galileo mission to the semiconductor industry, discusses his journey from early electronics hobbyist to chip designer at Texas Instruments and Signetics. He reflects on the evolution of embedded systems, comparing legacy architectures like the 8051 and 4-bit microcontrollers to modern ARM and FPGA technologies, while emphasizing the importance of continuous learning and professional resilience. Comer shares anecdotes about working on the Galileo probe's memory control system, the impact of alpha particles on space electronics, and the shifting landscape of major tech companies like HP and Intel.

    Careers & educationMicrocontrollers & dev platforms
    10 chapters

  • The Next Byte Podcast · 23m

    107. Mitigating High Emissions of Autonomous Cars

    Hosts Daniel and Forbode discuss research from MIT’s Laboratory for Information and Decision Systems, led by Sumia Sudakar, Vivian Sze, and Sertac Karaman, which estimates that autonomous vehicle computing could generate emissions equivalent to all global data centers. The team proposes a cap of 1.2 kilowatts per vehicle’s computing system, requiring hardware efficiency to double annually to maintain carbon neutrality. The hosts argue for specialized, in-house chip designs and software optimization, citing Google Pixel’s camera strategy as a successful model for maximizing value from fixed hardware rather than chasing the latest sensors.

    Energy & climate techMicrocontrollers & dev platforms
    7 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by EE Times Current. The episode itself belongs to its publisher.