English

Scientists demonstrate a new optical neural network training method that can crush electronic microprocessors

1224
2023-09-27 15:24:41
See translation

The current deep neural network system (such as ChatGPT) can quickly improve energy efficiency by 100 times in training, and "future improvements will greatly increase by several orders of magnitude. Scientists from MIT and other institutions have demonstrated a new optical neural network training method that can crush state-of-the-art electronic microprocessors.

Moreover, the computational density of the demonstrated system is about two orders of magnitude higher than that of Nvidia, Google, or Graphcore systems.

Basically, this means that the most advanced models can be trained with 100 times less energy and occupy less space at the same speed.

Artificial neural networks mimic the way biological brains process information. These artificial intelligence systems aim to learn, combine, and summarize information from big datasets, reshaping the field of information processing. Current applications include images, objects, speech recognition, games, medicine, and physical chemistry.

The current artificial intelligence model has reached hundreds of billions of artificial neurons, showing exponential growth and posing challenges to current hardware capabilities.

This paper demonstrates that optical neural network (ONN) methods with high clock speed, parallelism, and low loss data transmission can overcome current limitations.

Our technology opens up a path for large-scale optoelectronic processors to accelerate machine learning tasks from data centers to decentralized edge devices, "the paper wrote.

The ONN method is expected to alleviate the bottlenecks of traditional processors, such as the number of transistors, data mobility energy consumption, and semiconductor size. ONN uses light, which can carry a large amount of information simultaneously due to its wide bandwidth and low data transmission loss. In addition, many photonic circuits can be integrated to expand the system.

In order to move light for calculation, the team led by MIT utilized many laser beams, which were described as "using mass-produced micrometer scale vertical cavity surface emitting lasers for neuron coding".

The researchers explained, "Our scheme is similar to the 'axon synapse dendrite' structure in biological neurons
They believe that the demonstrated system can be expanded through mature wafer level manufacturing processes and photon integration.

Dirk Englund, Associate Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology and the head of this work, explained to SciTechDaily that the size of models such as ChatGPT is limited by the capabilities of today's supercomputers. Therefore, training larger models is not economically feasible.

He claimed, "Our new technology can make it possible to cross machine learning models, otherwise it would not be possible in the near future.

This paper titled "Deep Learning Using Coherent VCSEL Neural Networks" was published by a large team of scientists. This work has received support from the Army Research Office, NTT Research, and NTT Netcast Awards, as well as financial support from the Volkswagen Foundation. The three researchers of the team have applied for patents related to this technology.

Source: Laser Network

Related Recommendations
  • IPG Japan office and technical center officially opened

    Recently, IPG Photonics, a leading company in the global fiber laser field, announced the official opening of its new office and central technology center in Japan, marking a solid step in the technology giant's strategic deployment in the Asia Pacific region.The opening of this new office not only demonstrates IPG Photonics' high regard for Japan and the entire Asia Pacific market, but also indic...

    2024-07-15
    See translation
  • Progress in the study of ultrafast electron dynamics using short light pulses

    When electrons move in molecules or semiconductors, their time scale is unimaginably short. The Swedish German team, including Dr. Jan Vogelsang from the University of Oldenburg, has made significant progress in these ultrafast processes: researchers are able to track the dynamics of electrons released on the surface of zinc oxide crystals using laser pulses with nanoscale spatial resolution and p...

    2024-01-08
    See translation
  • Micro active vortex laser

    Recently, Dong Yibo, from the Photonic Chip Research Institute of Shanghai University of Technology, published his research findings titled "Nanoprinted Diffractive Layer Integrated Vertical Cavity Surface Emitting Vortex Lasers with Scalable Topological Charge" as the first author in the internationally renowned journal Nano Letters.This achievement was jointly completed by the team of academicia...

    2023-10-24
    See translation
  • A research team at City University of Hong Kong has developed a multispectral, ultra-low dose photoacoustic microscope system

    Optical resolution "photoacoustic microscope is a new biomedical imaging technology, which can be used in the research of cancer, diabetes, stroke and other diseases. However, insufficient sensitivity has always been a long-term obstacle to its wider application.According to Maims Consulting, a research team from City University of Hong Kong (CityU) has recently developed a multispectral, ultra-lo...

    2023-09-21
    See translation
  • New photon avalanche nanoparticles may usher in the next generation of optical computers

    A research team led by Lawrence Berkeley National Laboratory (Berkeley Lab), Columbia University, and Autonomous University of Madrid has successfully developed a novel optical computing material using photon avalanche nanoparticles. This breakthrough achievement was recently published in the journal Nature Photonics, paving the way for the manufacture of optical memory and transistors at the nano...

    02-28
    See translation