Biology subjects

Saxe, A.

Publications and source records attributed to Saxe, A..

3 recordsLinked to original sources

bioRxiv2021

Neural knowledge assembly in humans and deep networks

Human understanding of the world can change rapidly when new information comes to light, such as when a plot twist occurs in a work of fiction. This flexible "knowledge assembly" requires few-shot reorganisation of neural codes for relations among objects and events. However, existing computational theories are largely silent about how this could occur. Here, participants learned a transitive ordering among novel objects within two distinct contexts, before exposure to new knowledge that revealed how they were linked. BOLD signals in dorsal frontoparietal cortical areas revealed that objects were rapidly and dramatically rearranged on the neural manifold after minimal exposure to linking information. We then adapt stochastic online gradient descent to permit similar rapid knowledge assembly in a neural network model.

neuroscience↗

bioRxiv2021

Organizing memories for generalization in complementary learning systems

Memorization and generalization are complementary cognitive processes that jointly promote adaptive behavior. For example, animals should memorize a safe route to a water source and generalize to features that allow them to find new water sources, without expecting new paths to exactly resemble previous ones. Memory aids generalization by allowing the brain to extract general patterns from specific instances that were spread across time, such as when humans progressively build semantic knowledge from episodic memories. This cognitive process depends on the neural mechanisms of systems consolidation, whereby hippocampal-neocortical interactions gradually construct neocortical memory traces by consolidating hippocampal precursors. However, recent data suggest that systems consolidation only applies to a subset of hippocampal memories; why certain memories consolidate more than others remains unclear. Here we introduce a novel neural network formalization of systems consolidation that highlights an overlooked tension between neocortical memory transfer and generalization, and we resolve this tension by postulating that memories only consolidate when it aids generalization. We specifically show that unregulated memory transfer can be detrimental to generalization in unpredictable environments, whereas optimizing systems consolidation for generalization generates a high-fidelity, dual-system network supporting both memory and generalization. This theory of generalization-optimized systems consolidation produces a neural network that transfers some memory components to the neocortex and leaves others dependent on the hippocampus. It thus provides a normative principle for reconceptualizing numerous puzzling observations in the field and provides new insight into how adaptive behavior benefits from complementary learning systems specialized for memorization and generalization.

neuroscience↗

bioRxiv2021

Rich and lazy learning of task representations in brains and neural networks

How do neural populations code for multiple, potentially conflicting tasks? Here, we used computational simulations involving neural networks to define "lazy" and "rich" coding solutions to this multitasking problem, which trade off learning speed for robustness. During lazy learning the input dimensionality is expanded by random projections to the network hidden layer, whereas in rich learning hidden units acquire structured representations that privilege relevant over irrelevant features. For context-dependent decision-making, one rich solution is to project task representations onto low-dimensional and orthogonal manifolds. Using behavioural testing and neuroimaging in humans, and analysis of neural signals from macaque prefrontal cortex, we report evidence for neural coding patterns in biological brains whose dimensionality and neural geometry are consistent with the rich learning regime.

neuroscience↗