An Implicit Memory-Based Method for Supervised Pattern Recognition

Algorithm 1

Implicit Recognition Model.

	input: is a training instance set and is the known label of ; , labeled , is an approximator and can be trained; ; is a testing instance.
	Cognitive Process
(1)	for each do
(2)	repeat
(3)	Observing a signal in ;
(4)	Training the prediction function to predict the masked parts of the signal;
(5)	until For each signal and any , .
(6)	end for
(7)	return Prediction functions labeled with respectively.
	Recognition Process
(1)	for each do
(2)	Estimating the similarity between testing signal and instance set ;
(3)	Let ;
(4)	end for
(5)	Assigning the image to the class of its highest similarity;
(6)	;
(7)	return.