Wavelength-multiplexed Delayed Inputs for Memory Enhancement of Microring-based Reservoir Computing

This work addresses memory limitations in reservoir computing for optical computing applications, but it appears incremental as it builds on existing methods with specific hardware optimizations.

The authors tackled the problem of memory-demanding tasks like time-series prediction by developing a silicon microring-based reservoir computing scheme that uses parallel delayed inputs and wavelength division multiplexing, achieving good performance without external optical feedback.