The blossoming of artificial intelligence technologies demands greater computational power. Optical computing, with its ultra-fast operation and inherent parallel processing capabilities, is considered promising for high-performance computing tasks. However, there is no evidence that optical computing outperforms classical neural networks in machine learning tasks. In this work, we explored this issue by evaluating the expressive and trainable capabilities of optic propagation processes. This evaluation was based on the Fisher information and the effective dimension theories. Practical machine learning experiments were conducted to validate our theoretical findings. The results revealed that optical-propagation-based computing exhibited a superior effective dimension and quicker training speed than classical feedforward neural networks, underscoring optical computing’s benefits in machine learning applications.
The all-optical/optoelectronic hybrid intelligent computing with high speed and low power consumption characteristics has emerged as a promising solution for the bottleneck of Moore's Law. However, the low deployment flexibility and low integration limit the further development of optical computing systems, and it is urgent to match the new artificial intelligence theory with the physical characteristics of light itself. In this paper, we proposed a reconfigurable computing architecture based on planar optical waveguide, called wave network, which can be used for hardware development of optical switches, optical logic devices or neural networks. Inspired by the neuronal perceptron model, the elements of wave network are analyzed from the point of view of basic pixel units. Wave network has the processing capability of pulse input and time sequence input. To achieve low-power weight mapping, non-volatile materials with CMOS compatiblity characteristics are introduced above the waveguide (etc. phase change materials, multiferroics materials, and resistive switching materials). This hybrid structure provides a large-scale, low-cost, rapid development platform for the new photonic computing paradigm. And the simulation results show that the properties of the waveguide remain stable at different geometric scales and have the potential of scaling down.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.