Computational Neurophysics group, Institute of Neurosciences and medicine, Jülich, Germany

Sequence learning in brain-inspired computing systems

Sequence learning is one of the universal computations performed by the vast but homogeneous network of the mammalian brain. The abstract Hierarchical Temporal Memory (HTM) algorithm accomplishes this form of computation. We present first steps of a spiking neuronal network realization compatible with the biological constraints. Nevertheless, rapid and energy efficient execution of brain-inspired algorithms requires corresponding hardware and software. The second part of the talk therefore evaluates the progress of large-scale neuromorphic computing systems.

Bouhadjar Y, Wouters DJ, Diesmann M, Tetzlaff T (2022) Sequence learning, prediction, and replay in networks of spiking neurons. PLoS Comput Biol 18(6):e1010233

Kurth AC, Senk J, Terhorst D, Finnerty J, Diesmann M (2022) Sub-realtime simulation of a neuronal network of natural density. Neuromorph Comput Eng 2:021001

Invited by Andrew Davison