Analog readout for optical reservoir computers
This work addresses a major limitation for the future development of hardware reservoir computers, enabling real-time operation, though it appears incremental as it focuses on improving an existing bottleneck.
The authors tackled the performance bottleneck of slow digital postprocessing in hardware reservoir computers by designing an analog readout for time-multiplexed optoelectronic systems, achieving better performance than non-reservoir methods on a standard benchmark task.
Reservoir computing is a new, powerful and flexible machine learning technique that is easily implemented in hardware. Recently, by using a time-multiplexed architecture, hardware reservoir computers have reached performance comparable to digital implementations. Operating speeds allowing for real time information operation have been reached using optoelectronic systems. At present the main performance bottleneck is the readout layer which uses slow, digital postprocessing. We have designed an analog readout suitable for time-multiplexed optoelectronic reservoir computers, capable of working in real time. The readout has been built and tested experimentally on a standard benchmark task. Its performance is better than non-reservoir methods, with ample room for further improvement. The present work thereby overcomes one of the major limitations for the future development of hardware reservoir computers.