PANN™ training algorithm
- Correction of the Omega-Neuron weights, i.e., training is not accomplished by gradually correcting weight values for one image after another (gradual gradient descent), but with a single-step operation of error compensation during the retrograde signal. This takes into account only the information received by the neuron from its synapses during training.
- Training the PANN™ for the next image does not depend on its training for the previous image. Complete compensation of training error is provided for each image used in training.
- Complex calculations are not necessary.
- Multiple repeating calculations are not necessary.
- In its basic variant, PANN™ provides correction of all weights that contribute to the error on a particular neuron; correction is provided by using the same value for all active weights.
- In its basic variant, PANN™ has no activation function (or has a linear activation function), which drastically simplifies and reduces calculations.
- Calculations and weight correction may be performed using matrix algebra.
These advantages enable a thousand-fold increase in network training speed.