Complex computation from developmental priors.

Machine learning (ML) models have long overlooked innateness: how strong pressures for survival lead to the encoding of complex behaviors in the nascent wiring of a brain. Here, we derive a neurodevelopmental encoding of artificial neural networks that considers the weight matrix of a neural network to be emergent from well-studied rules of neuronal compatibility. Rather than updating the network's weights directly, we improve task fitness by updating the neurons' wiring rules, thereby mirroring evolutionary selection on brain development. We find that our model (1) provides sufficient representational power for high accuracy on ML benchmarks while also compressing parameter count, and (2) can act as a regularizer, selecting simple circuits that provide stable and adaptive performance on metalearning tasks. In summary, by introducing neurodevelopmental considerations into ML frameworks, we not only model the emergence of innate behaviors, but also define a discovery process for structures that promote complex computations.

Neural heterogeneity promotes robust learning.

The brain is a hugely diverse, heterogeneous structure. Whether or not heterogeneity at the neural level plays a functional role remains unclear, and has been relatively little explored in models which are often highly homogeneous. We compared the performance of spiking neural networks trained to carry out tasks of real-world difficulty, with varying degrees of heterogeneity, and found that heterogeneity substantially improved task performance. Learning with heterogeneity was more stable and robust, particularly for tasks with a rich temporal structure. In addition, the distribution of neuronal parameters in the trained networks is similar to those observed experimentally. We suggest that the heterogeneity observed in the brain may be more than just the byproduct of noisy processes, but rather may serve an active and important role in allowing animals to learn in changing environments.