. Author manuscript; available in PMC: 2016 Sep 1.

Published in final edited form as: Brain Imaging Behav. 2016 Sep;10(3):818–828. doi: 10.1007/s11682-015-9430-4

Algorithm 1.

Pseudo code of the proposed method

Input: X⁽¹⁾ ∈ ℝ^d^×ⁿ, X⁽²⁾ ∈ ℝ^d^×ⁿ, Y ∈ ℝ^c^×ⁿ, β, γ

Output: W

1 Conduct CCA to obtain (B̂⁽¹⁾, B̂⁽²⁾) and λ₁, …, λ_d

2 Generate canonical representations by Eq. 3: Z = [Z⁽¹⁾; Z⁽²⁾] ∈ ℝ²^d^×ⁿ

3 Define the canonical regularizer according to Eq. 6

4 Optimize Eq. 7 via Algorithm 2 in A