AI-driven modelling for rapid and robust cortex-wide credit assignment


Learning requires the brain to assign credit to billions of synapses. How the brain achieves this feat is one of the unsolved mysteries in neuroscience. Recently, inspired by deep learning, we have introduced a novel model of hierarchical credit assignment in cortico-cortical networks. Our model combines synaptic, sub-cellular, cellular, microcircuit and cortico-cortical computations to enable error-driven learning of challenging tasks. I will show that in contrast to previous work our model (i) is consistent with experimental observations, (ii) provides rapid credit assignment across multiple cortical areas and (iii) does not require a multi-phase learning process.

Experimental evidence suggests that neuromodulation also plays a key role in controlling learning. Inspire by deep learning we model cholinergic neuromodulation as an adaptive learning system. In our model cholinergic neuromodulation democratizes learning by continuously shifting learning to different neuronal populations. I will show that such distributed learning has two key consequences: (i) greatly speeds up learning and (ii) produces a more distributed task-encoding. Importantly, more distributed representations result in networks that are more robust to perturbations (e.g. cell death), thereby providing the first theoretical explanation of why Cholinergic deficits are commonly associated with dementia, aging and injury.
In summary, our AI-driven modelling is opening the window to a new understanding of learning in the brain with important implications for health and disease.


Rui leads the Neural & Machine Learning group at the University of Bristol. The group builds biologically-constrained AI-driven models to transform our understanding of how the brain learns. The group is funded by the EPSRC, BBSRC, Wellcome Trust, MRC and a recently awarded ERC grant to pursue this research program. Before starting his group Rui did postdoctoral research in computational neuroscience & machine learning at the University of Oxford and briefly at the University of Bern. ​Previously, he completed his PhD in 2015 at the University of Edinburgh (UK) as part of the Institute for Adaptive and Neural Computation with Mark van Rossum and P. Jesper Sjöström (funded by a FCT PhD grant). During that time Rui was also a visiting PhD student at University College London (UK) and McGill University (Canada). Before that Rui studied computer science at the University of Coimbra (Portugal). Rui used to organise the Oxford NeuroTheory Forum (2014-2017).