Computation across scales in brain and beyond

The brain operates across multiple interacting scales from the level of ion channels in individual neurons to the population of interacting neurons and even beyond, in connection with the rest of the body. Hence, to fully understand the mechanisms underlying brain function and ultimately behavior, we need to study both neural dynamics and neural computations across multiple scales of organization. Last year, we organized a successful satellite workshop addressing the multi-scale dynamics of the brain (“Multi-modal understanding of brain and behavior”). This year, we propose a complementary workshop to discuss brain computations across different scales.

In this workshop, we bring together a broad group of scientists in computational and experimental neuroscience as well as artificial intelligence (AI), studying distinct types of computation across different scales. Our speakers will discuss how they apply various computational frameworks including efficient coding, reinforcement learning, Bayesian inference, information theory, and deep learning to formalize computational and learning algorithms not only from the level of single neurons to interacting brain areas, but also considering computations with non-neuron cell types (e.g., astrocytes), and eventually brain-body interactions. We aim to stimulate discussions on the benefits and challenges of different computational frameworks, how they can facilitate a multi-scale understanding of brain computation, and their potential applications to advancing AI algorithms. We believe through this exchange of ideas, we can find new paths toward systemizing an approach to uncover a “hierarchy” of computation in the brain. Moreover, the complementary perspectives of our speakers synergized with discussions in our workshop would promote new research directions and collaborations that will broaden the extent of multi-scale brain research to gain a more holistic understanding of brain computation and cognition.