Dscc 435 Opt For Ml

Examples of ADMM https://jiaming-liang.github.io/OPTML.html.

Convergence analysis and constrained optimization https://jiaming-liang.github.io/OPTML.html.

Course logistics and introduction to optimization https://jiaming-liang.github.io/OPTML.html.

Relative smooth and strongly convex. Examples. https://jiaming-liang.github.io/OPTML.html.

A unified treatment of three variants https://jiaming-liang.github.io/OPTML.html.

Projection, convergence analysis, and subgradient https://jiaming-liang.github.io/OPTML.html.

Primal gradient and dual averaging methods https://jiaming-liang.github.io/OPTML.html.

Complexity notion and gradient method https://jiaming-liang.github.io/OPTML.html.

High probability result of stochastic subgradient method under sub-Gaussian assumption ...

Localization and convergence analysis https://jiaming-liang.github.io/OPTML.html.

Nesterov's smoothing technique https://jiaming-liang.github.io/OPTML.html.

Proximal mapping, Moreau envelope, and composite optimization https://jiaming-liang.github.io/OPTML.html.

Pointwise and ergodic convergence, saddle point problem, and Chabolle-Pock method.

Chambolle-Pock as IPP, dual ascent, method of multipliers, ADMM as IPP.

Understanding Frank-Wolfe as accelerated gradient without acceleration. IPP framework convergence and examples.

Monotone operator and generalized IPP framework convergence.

Geometric interpretation, convex analysis, and convergence analysis https://jiaming-liang.github.io/OPTML.html.

Connection between sampling and stochastic optimization. Approximation error in SAA.