Partial recurrence can enable robust and efficient computation

Abstract

Neural circuits are sparse and bidirectional. Meaning that signals flow from early sensory areas to later regions and back. Yet, between connected areas there exist some but not all pathways. How does this structure, somewhere between feedforward and fully recurrent, shape circuit function? To address this question, we designed a recurrent neural network model in which a set of weight matrices (i.e. pathways) can be combined to generate every network structure between feedforward and fully recurrent. We term these architectures partially recurrent neural networks (pRNNs). We trained over 25,000 pRNNs on a novel set of reinforcement learning tasks, designed to mimic multisensory navigation, and compared their performance across multiple functional metrics. Our findings reveal three key insights. First, in dense-cue environments, most pRNN architectures match or exceed the task performance, learning speed or robustness of fully recurrent networks, despite using as few as one quarter the number of parameters; in sparse-cue environments, many match but a substantial fraction underperform. These results demonstrate that partial recurrence can enable energy efficient, yet performant solutions. Second, each pathway’s functional impact is both task and circuit dependent. For instance, feedback connections enhance robustness to noise in some, but not all contexts. Third, different pRNN architectures learn solutions with distinct input sensitivities and memory dynamics, and these computational traits help to explain their functional capabilities. Overall, our results demonstrate that partial recurrence can enable robust and efficient computation - a finding that may help to explain why neural circuits are sparse and bidirectional, and shows how these principles can inform the design of artificial systems.