Online Multikernel Learning Based on a Triple-Norm Regularizer for Semantic Image Classification

Algorithm 1

Pseudocode of the TripleReg-MKL algorithm. Parameters   and   are initially set to zero as online learning algorithms usually do [31]. It is intuitive that model parameters and are empty at the starting time when no sample arrives.

TripleReg-MKL

() Input ;

() Initialize  

() For do

()  Receive a new instance randomly from

()  Make a prediction

()  Obtain the correct label

()  Compute inference label, that is,

()   if  

()         

()  ;

()  ;  

()  end if

() end for