Gradient of V_t calc
ofTarradiddle opened this issue · comments
Should Vnew = V_t + self.step[t] * (torch.mm(U_t.t(),E_t).t() @ Ukernel) be (torch.mm(E_t,U_t) @ Ukernel)?
Learned Robust PCA: A Scalable Deep Unfolding Approach for High-Dimensional Outlier Detection
ofTarradiddle opened this issue · comments
Should Vnew = V_t + self.step[t] * (torch.mm(U_t.t(),E_t).t() @ Ukernel) be (torch.mm(E_t,U_t) @ Ukernel)?