HW1 constrained optimization

Moderator: Statistisches Maschinelles Lernen

MFB
Erstie
Erstie
Beiträge: 21
Registriert: 3. Okt 2013 17:10

HW1 constrained optimization

Beitrag von MFB »

Hi, I have to admit I am struggling a bit with the optimization problem.
See images. So I got the partial derivatives and the dual. I can now take the derivatives of the lagrange multipliers, equal them to 0 and solve them for the lambdas, which get plugged into the partial derivatives of the probabilities. My problem is neither the primal nor dual are pleasant to solve analytically
IMG_20180719_145705.jpg
IMG_20180719_145705.jpg (44.64 KiB) 398 mal betrachtet
IMG_20180719_145711.jpg
IMG_20180719_145711.jpg (43.28 KiB) 398 mal betrachtet
Dateianhänge
IMG_20180719_145705.jpg
IMG_20180719_145705.jpg (44.64 KiB) 399 mal betrachtet

Zurück zu „Statistisches Maschinelles Lernen“