The OJA neural network, also known as Oja's rule neural network, is a type of artificial neural network used for unsupervised learning in dimension reduction and principal component analysis problems.
It was developed by Finnish mathematician Erkki Oja in 1982 and is based on a learning algorithm that allows the neural network to find the principal directions of the input features and reduce the dimension of the data. Compared to other dimension reduction methods, the OJA neural network is better able to handle highly correlated and non-linear data.
The operation of the OJA neural network is based on adjusting the synaptic weights of the network so that the output neuron responds selectively to specific input patterns. This is achieved by iteratively calculating the synaptic weights to maximise the correlation between the inputs and the output neuron.
Nowadays digital transformation is key in any type of business. The 40% of Spanish companies will not exist in its current form in the next few [...]
Read More »How is artificial intelligence helping us? Artificial intelligence (AI) has gone from being the stuff of science fiction movies to a [...]
Read More »One of the decisions faced by a company that needs an IT infrastructure is the choice of where to locate this infrastructure and where to install it.
Read More »The term artificial intelligence (AI) is nowadays, but it was invented in 1956 by John McCarthy, Marvin Minsky and Claude Shannon in the famous [...]
Read More »
