Topic: error backpropagation

Fast cortical processing is required in many scenarios where both sensory input and the corresponding cognitive responses change rapidly. Therefore, efficient learning in such networks needs to tackle the issue of credit assignment continuously, in real time. Recent years have witnessed a surge of cortical learning models which address this question by approximating the error backpropagation algorithm. However, all of these either require long relaxation phases following a change in sensory stimuli, which renders them unable to cope with the fast time scales imposed by, e.g., saccades, or impo

Effective learning in neuronal networks requires the adaptation of individual synapses given their relative contribution to solving a task. However, physical neuronal systems -- whether biological or artificial -- are constrained by spatio-temporal locality. In other words, synapses can only use information available at their physical location and at the present moment in time. How such networks can perform efficient credit assignment, remains, to a large extent, an open question. In machine learning (ML), the answer is almost universally given by the error backpropagation algorithm, through b

To resolve the fundamental question of how the mammalian neocortex learns to create complex hierarchical representations, two main competing theories that intersect ideas from neuroscience and machine learning are currently debated. The first theory draws inspiration from artificial neural networks, suggesting the brain approximates error backpropagation (BP), where neurons adjust synapses to minimize a backpropagated error. The second approach, target learning (TL), proposes that neurons learn by reducing feedback needed to achieve desired activity patterns. While both theories are computati

Amyotrophic Lateral Sclerosis (ALS) is a progressive neurological disorder characterized by the gradual degeneration of cells in the spinal cord, brainstem, and motor cortex, leading to manifestations such as weakness, muscle atrophy, and spasticity (Elisen, 2009). A comprehensive understanding of the disease process is currently lacking, making preclinical ALS models crucial to study the mechanisms of pathology insurgence, characterize the disease and test therapeutics approaches. Pathological in vitro models emerged as viable alternative to in vivo studies, consolidating a non-invasive appro