Neural astrocyte networks accumulate evidence to adapt to changing contexts
A neural-astrocyte architecture implements a hybrid automaton for evidence accumulation
Neural and Evolutionary Computing
Summary
This paper explores how brain cells called astrocytes, which support neurons, might help neural networks learn and adapt to changing situations. The authors created a model combining neurons and astrocytes that can remember and respond to shifts in rules based on rewards. Their model shows astrocytes can build up information over time and change brain activity patterns to switch between different contexts. This process mimics how some brain systems accumulate evidence until they decide to change behavior.
What this means in practice
- •For ai system designers: Create reinforcement learning models that better handle changing environments by integrating astrocyte-inspired dynamics for context inference.
- •For neuromorphic hardware engineers: Incorporate astrocyte-like mechanisms to design chips that switch neural processing states based on accumulating environmental evidence.
Authors
Giacomo Vedovati, Ilya E. Monosov, Thomas J. Papouin, ShiNung Ching
Abstract
Astrocytes are non-neuronal glial cells that are receiving widespread attention due to their emerging role in neural computation. In this paper, we propose and study dynamical mechanisms by which astrocytes may augment the ability of neural networks to infer context in reinforcement learning (RL) settings. We construct a biologically inspired, two-level dynamical neural-astrocyte network with distinct spatial and temporal organization. We train this model on a hierarchical multi-context task that requires the agent to infer changes in latent task rules based on derived rewards. We find that in this setting, astrocytes enable evidence accumulation of changes in context and subsequent context-specific modulation of neural dynamics. We show that these functions are implemented via two dynamical mechanisms: (i) reward-induced bifurcations that relocate an asymptotically stable attractor into different, context-specific regions of state space, and (ii) the relative shallowness of these attractors, mediated by the entropy of the environment, giving rise to behavioral stickiness. Together, these mechanisms amount to a hybrid automaton, in which uncertainty accumulates until, eventually, the neural dynamics are switched to a new context. This model provides a neuro-dynamic schema, compatible with neural-astrocyte biology and prior empirical observations, for how astrocytes may integrate information from the periphery and drive contextual changes in neural circuits.