Researchers have shown that it is possible to train artificial neural networks directly on an optical chip. The research demonstrates that an optical circuit can perform a critical function of an electronics-based artificial neural network, and that it could lead to less expensive, faster, and more energy-efficient ways to perform tasks such as speech or image recognition. Researchers have shown a neural network can be trained using an optical circuit (blue rectangle). In the full network, there would be several of these linked together. The laser inputs (green) encode information that is carried through the chip by optical waveguides (black). The chip performs operations crucial to the artificial neural network using tunable beamsplitters, which are represented by the curved sections in the waveguides. These sections couple two adjacent waveguides together and are tuned by adjusting the settings of optical phase shifters (red and blue glowing objects), which act like ‘knobs’ that can be adjusted during training to perform a given task. Courtesy of Tyler W. Hughes, Stanford University. The optical chip designed by the researchers from Stanford University replicates the way that conventional computers train neural networks. The new training protocol operates on optical circuits with tunable beamsplitters that are adjusted by changing the settings of optical phase-shifters. Laser beams encoding information to be processed are fired into the optical circuit and carried by optical waveguides through the beamsplitters, which are adjusted like knobs to train the neural network algorithms. The laser is first fed through the optical circuit. Upon exiting the device, the difference from the expected outcome is calculated. This information is then used to generate a new light signal, which is sent back through the optical network in the opposite direction. By measuring the optical intensity around each beamsplitter during this process, the researchers showed how to detect, in parallel, how the neural network performance will change with respect to each beamsplitter’s setting. The phase-shifter settings can be changed based on this information, and the process can be repeated until the neural network produces the desired outcome. The researchers tested their training technique with optical simulations by teaching an algorithm to perform complicated functions, such as picking out complex features within a set of points. They found that the optical implementation performed similarly to a conventional computer. “Our work demonstrates that you can use the laws of physics to implement computer science algorithms,” said researcher Shanhui Fan. “By training these networks in the optical domain, it shows that optical neural network systems could be built to carry out certain functionalities using optics alone.” Although optical artificial neural networks were recently demonstrated experimentally, the training step was performed using a model on a traditional digital computer, and the final settings were then imported into the optical circuit. The Stanford team uses a method for training these networks directly in the device by implementing an optical analogue of the “backpropagation” algorithm, which is the standard way to train conventional neural networks. “Using a physical device rather than a computer model for training makes the process more accurate,” said researcher Tyler W. Hughes. “Also, because the training step is a very computationally expensive part of the implementation of the neural network, performing this step optically is key to improving the computational efficiency, speed, and power consumption of artificial networks.” The researchers plan to further optimize the system and want to use it to implement a practical application of a neural network task. The general approach they designed could be used with various neural network architectures and for other applications, such as reconfigurable optics. “Using an optical chip to perform neural network computations more efficiently than is possible with digital computers could allow more complex problems to be solved,” Fan said. “This would enhance the capability of artificial neural networks to perform tasks required for self-driving cars or to formulate an appropriate response to a spoken question, for example. It could also improve our lives in ways we can’t imagine now.” The research was published in Optica, a publication of OSA, The Optical Society (doi:10.1364/OPTICA.5.000864).