Règle de Hebb | EPFL Graph Search

La règle de Hebb, théorie de Hebb, postulat de Hebb ou théorie des assemblées de neurones a été établie par Donald Hebb en 1949. Elle est à la fois utilisée comme hypothèse en neurosciences et comme concept dans les réseaux neuronaux en mathématiques. En 1950, un manuscrit de Sigmund Freud datant de 1895 fut publié qui attestait que cette théorie avait déjà été formulée avant Hebb. Cette théorie est souvent résumée par la formule : () C'est une règle d'apprentissage des réseaux de neurones artificiels dans le contexte de l'étude d'assemblées de neurones. Elle est basée sur une hypothèse scientifique en neurosciences. Elle décrit les changements d'adaptation neuronale dans le cerveau ou dans un réseau de neurones pendant un processus d'apprentissage. Elle décrit un mécanisme basique de plasticité synaptique dans laquelle l'efficacité synaptique augmente lors d'une stimulation présynaptique répétée et persistante de la cellule postsynaptique. Cette règle suggère que lorsque deux

Reverse engineering the motor control system

Berat Denizdurduran

Mammalian motor control is implemented by a combination of different networks and system, working coherently to plan the movement of the body or a limb and to execute this movement a dynamical environment. While it is believed that complex, voluntary movements are planned in the motor areas of the cortex, the execution of the movement is controlled by a combination of cortical and cerebellar networks together with central pattern generators and reflex circuits in the spinal cord. In this thesis, we propose an abstract model that captures the basic properties of mammalian motor control, using the example of two different movements: arm movements as well as locomotion. Our model consists of three parts: first a high-level network, that learns movements, using a combination of actor-critic based reinforcement learning and Optimal Control theory. The second part corresponds to spinal reflex circuits that execute basic movements of the musculo-skeletal system. They are modeled by a simple neural network, that learns the dynamic properties of the muscoloskeletal system by the mechanism of spontaneous motor activity (muscle twithing) combined with a Hebbian learning rule. We demonstrate that this network can learn the antagonistic control of joint movements. The final part of the model is a cerebellar network, that translates a complex movement trajectory, such as reaching, and activates the spinal reflex circuits to execute the movement. The mapping between the cerebellar neurons and the spinal reflex circuits are trained with artificial neural networks. Using a musculoskeletal arm model, we demonstrate that the proposed neural motor control model can generate the movement to arbitrary goal position.

EPFL2019

Dendritic Voltage Recordings Explain Paradoxical Synaptic Plasticity: A Modeling Study

Wulfram Gerstner, Laureline Logiaco, Claire Meissner-Bernard, Matthias Chinyen Tsai

Experiments have shown that the same stimulation pattern that causes Long-Term Potentiation in proximal synapses, will induce Long-Term Depression in distal ones. In order to understand these, and other, surprising observations we use a phenomenological model of Hebbian plasticity at the location of the synapse. Our model describes the Hebbian condition of joint activity of pre- and postsynaptic neurons in a compact form as the interaction of the glutamate trace left by a presynaptic spike with the time course of the postsynaptic voltage. Instead of simulating the voltage, we test the model using experimentally recorded dendritic voltage traces in hippocampus and neocortex. We find that the time course of the voltage in the neighborhood of a stimulated synapse is a reliable predictor of whether a stimulated synapse undergoes potentiation, depression, or no change. Our computational model can explain the existence of different -at first glance seemingly paradoxical- outcomes of synaptic potentiation and depression experiments depending on the dendritic location of the synapse and the frequency or timing of the stimulation.

FRONTIERS RESEARCH FOUNDATION2020

Biologically plausible unsupervised learning in shallow and deep neural networks

Bernd Albert Illing

The way our brain learns to disentangle complex signals into unambiguous concepts is fascinating but remains largely unknown. There is evidence, however, that hierarchical neural representations play a key role in the cortex. This thesis investigates biologically plausible models of unsupervised learning of hierarchical representations as found in the brain and modern computer vision models. We use computational modeling to address three main questions at the intersection of artificial intelligence (AI) and computational neuroscience.The first question is: What are useful neural representations and when are deep hierarchical representations needed? We approach this point with a systematic study of biologically plausible unsupervised feature learning in a shallow 2-layer networks on digit (MNIST) and object (CIFAR10) classification. Surprisingly, random features support high performance, especially for large hidden layers. When combined with localized receptive fields, random feature networks approach the performance of supervised backpropagation on MNIST, but not on CIFAR10. We suggest that future models of biologically plausible learning should outperform such random feature benchmarks on MNIST, or that such models should be evaluated in different ways.The second question is: How can hierarchical representations be learned with mechanisms supported by neuroscientific evidence? We cover this question by proposing a unifying Hebbian model, inspired by common models of V1 simple and complex cells based on unsupervised sparse coding and temporal invariance learning. In shallow 2-layer networks, our model reproduces learning of simple and complex cell receptive fields, as found in V1. In deeper networks, we stack multiple layers of Hebbian learning but find that it does not yield hierarchical representations of increasing usefulness. From this, we hypothesise that standard Hebbian rules are too constrained to build increasingly useful representations, as observed in higher areas of the visual cortex or deep artificial neural networks.The third question is: Can AI inspire learning models that build deep representations and are still biologically plausible? We address this question by proposing a learning rule that takes inspiration from neuroscience and recent advances in self-supervised deep learning. The proposed rule is Hebbian, i.e. only depends on pre- and post-synaptic neuronal activity, but includes additional local factors, namely predictive dendritic input and widely broadcasted modulation factors. Algorithmically, this rule applies self-supervised contrastive predictive learning to a causal, biological setting using saccades. We find that networks trained with this generalised Hebbian rule build deep hierarchical representations of images, speech and video.We see our modeling as a potential starting point for both, new hypotheses, that can be tested experimentally, and novel AI models that could benefit from added biological realism.

EPFL2021

Biologically plausible unsupervised learning in shallow and deep neural networks

Bernd Albert Illing

EPFL2021