Imagine the cerebellum as a powerful kernel machine, with each granule cell serving as a unique basis function. Just like how a kernel machine operates by implicitly using high-dimensional feature spaces, the cerebellum can map sensorimotor information to a vast and potentially infinite feature space. This expansion recoding in the granule cell layer provides significant computational advantages and plays a vital role in information processing. It is fascinating to consider that the generation of these kernel basis functions could be biologically plausible, as the core concept of a kernel machine involves memorizing crucial input patterns. Researchers have explored potential regimes for developing these kernels under limited resources and are investigating the benefits and drawbacks of each regime through simulations. To delve deeper into this groundbreaking research on the cerebellum’s kernel machine and its implications, read the full article!

Sensorimotor information provided by mossy fibers (MF) is mapped to high-dimensional space by a huge number of granule cells (GrC) in the cerebellar cortex’s input layer. Significant studies have demonstrated the computational advantages and primary contributor of this expansion recoding. Here, we propose a novel perspective on the expansion recoding where each GrC serve as a kernel basis function, thereby the cerebellum can operate like a kernel machine that implicitly use high dimensional (even infinite) feature spaces. We highlight that the generation of kernel basis function is indeed biologically plausible scenario, considering that the key idea of kernel machine is to memorize important input patterns. We present potential regimes for developing kernels under constrained resources and discuss the advantages and disadvantages of each regime using various simulation settings.

Read Full Article (External Site)