Time-multiplexed Reservoir Computing with Quantum-Dot Lasers: Does more complexity lead to better performance?
Reservoir computing with optical devices offers an energy-efficient approach for time-series forecasting. Quantum dot lasers with feedback are modelled in this paper to explore the extent to which increased complexity in the charge carrier dynamics within the nanostructured semiconductor can enhance the prediction performance. By tuning the scattering interactions, the laser's dynamics and response time can be finely adjusted, allowing for a systematic investigation. It is found that both system response time and task requirements need to be considered to find optimal operation conditions. Further, lasers with pronounced relaxation oscillations outperform those with strongly damped dynamics, even if the underlying charge carrier dynamics is more complex. This demonstrates that optimal reservoir computing performance relies not only on internal complexity but also on the effective utilization of these dynamics through the output sampling process.