We show a close relationship between the Expectation - Maximization (EM) algorithm and direct optimization algorithms such as gradientbased methods for parameter learning. We iden...
Ruslan Salakhutdinov, Sam T. Roweis, Zoubin Ghahra...
Convergence of blind delayed source separation algorithms, which use constant learning rates, is known to be slow. We propose a fuzzy logic based approach to adaptively select the...
Subspace learning approaches have attracted much attention in academia recently. However, the classical batch algorithms no longer satisfy the applications on streaming data or la...
Jun Yan, Benyu Zhang, Shuicheng Yan, Qiang Yang, H...
Many settings of unsupervised learning can be viewed as quantization problems — the minimization of the expected quantization error subject to some restrictions. This allows the ...
Alex J. Smola, Robert C. Williamson, Sebastian Mik...
Temporal difference (TD) algorithms are attractive for reinforcement learning due to their ease-of-implementation and use of "bootstrapped" return estimates to make effi...