Hebbian learning

Artificial Neural Networks (ANNs) attempt to emulate their biological counterparts. McCulloch and Pitts (1943) proposed a simple model of a neuron, and Hebb (1949) described a technique which became known as Hebbian learning. Rosenblatt (1961), devised a single layer of neurons, called a Perceptron, that was used for optical pattern recognition. 

The so-called Hebbian learning rule (to honour Canadian neuropsychologist Donald Hebb, who proposed it in 1949) specifies how much the weight of the connection between two neurons should increase or decrease as a function of their activation. The rule states that the weight should increase whether the two interconnected neurons are active simultaneously (otherwise the weight should decrease). This rule is not used too frequently nowadays because it works well provided that all the input patterns are orthogonal or uncorrelated. 

Montague PR, Dayan P, Sejnowski TJ (1996) A framework for mesencephalic DA systems based on predictive Hebbian learning. J Neurosci 76 1936-1947. 

To test this idea and the conditions on how the system needs to be organized in terms of the connectivity, initially and after learning, we introduced a Hebbian learning rule for the synapses from KCs to eKCs ... 

In summary, the more detailed models reveal how the nervous system of the locust may implement the elements of the odor classification scheme with random connections using simple elements like mutual all-to-all inhibition and Hebbian learning. We have also seen that the implementation of classification with these simple ingredients automatically solves the additional task of detecting the cluster structure of the input pattern set. 

Experiment 1 compares Hebbian trained Hopfield networks with their equivalent HEDA models. The aim is to discover whether or not the HEDA model can achieve the capacity of the Hopfield network. Hopfield networks were trained on patterns using standard Hebbian learning, with one pattern at a time being added until the network s capacity was exceeded. At this point, the learned patterns were set to be the targets for the HEDA search using Eqs. 21 and 22 and the network s weights were reset. 

F. Sommer and G. Palm, Neural Networks, 12,281 (1999). Improved Bidirectional Retrieval of Sparse Patterns Stored by Hebbian Learning. 

The fast mode of task a22y must be different from the fast mode of task v22y. Here the non target stimuli must be remembered too but these can be connected to each other by Hebbian learning. So we meet the following search processes in the fast mode ofa22y ... 

Hebbian learning, of presynaptic and postsynaptic activation, 12-3 Heetderks, W.J., 34-3 Helfrich, W., 62-4 Helicotrema, 5-4 Hellebrandt, F.A., 76-20 Heller, L, 37-7 Helmholtz, H.L.F., 27-4 Helmholtz, H., 4-7-4-S Hematocytes, 1 -2 constituents, 1-2 fundamentals, 60-2 mechanics and deformability, 60-1-60-11 red cells, 60-3-60-7 stresses and strains in, 60-2 Hematocytopoiesis, 1-2 Hemidecussation, 4-3 Hemmert, W 63-9 Henson cells, in cochlear... 

The third approach is Hebbian learning, where learning is done locally by adjusting the weight based on the activities of the neurons. 

---