Sdrjan Ostojic, LNC2, Ecole Normale Supérieure
Minimal implementations of behavioral tasks using low-rank recurrent neural networks.
Large scale neural recordings have established that the transformation of sensory stimuli into motor outputs relies on low-dimensional dynamics at the population level, while individual neurons exhibit complex activity. Understanding how low-dimensional computations on mixed, distributed representations emerge from the structure of the recurrent connectivity and inputs to cortical networks is a major challenge. We have recently introduced a novel class of recurrent neural networks, in which the connectivity is a sum of a random part and a minimal, low-dimensional structure. In this presentation I will describe how in this class of networks low-dimensional dynamics can be directly predicted from connectivity. I will then show how this understanding can be leveraged to design minimal recurrent networks that perform specific computational tasks.