Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper investigates how interaction within coupled Echo State Networks (ESNs) can improve collective performance beyond simple aggregation. By training ESN pairs independently and then testing them jointly with and without coupled feedback, the authors demonstrate a "dyadic advantage" with pooled feedback, unlike simple aggregation which diverged. Furthermore, larger groups of smaller interacting ESNs achieved better accuracy and efficiency than smaller groups of larger ESNs for equivalent computation.
Interacting Echo State Networks outperform simple aggregation and achieve greater computational efficiency than larger independent models.
Simple aggregation of independent responses ("wisdom of crowds") generally improves collective performance through statistical facilitation, but interaction among members can sometimes hinder group outcomes. However, under certain circumstances, coordination studies show that interacting groups can outperform their individual members and nominal groups, and interaction can also be more efficient than independence. We present a reservoir computing model to investigate how interaction can enhance joint performance in terms of accuracy and efficiency. Echo state networks were trained independently to identify Japanese speakers from vowel recordings and tested jointly with and without coupled feedback. Model pairs reproduced the dyadic advantage reported by Bahrami et al. (2010) with interaction via pooled-feedback, but not when independent responses were aggregated together. Model dynamics revealed that interaction was beneficial whereas aggregated outputs were prone to diverging from runaway feedback over time. In terms of efficiency, larger groups of smaller interacting models outperformed smaller groups of larger models while equating for total amount of group computation. Our results offer a formal framework for explaining when and why interaction enhances collective intelligence.