A Riemannian Limited-memory BFGS Algorithm for Computing the Matrix Geometric Mean.

Xinru Yuan Wen Huang Pierre-Antoine Absil Kyle A. Gallivan

Published in: ICCS (2016)

Keyphrases

limited memory
quasi newton method
convergence rate
memory space
learning algorithm
objective function
high accuracy
simulated annealing
real time
computational complexity
reinforcement learning
optimal solution
np hard
probabilistic inference
high efficiency