ขนาดวิดีโอ: 1280 X 720853 X 480640 X 360
แสดงแผงควบคุมโปรแกรมเล่น
เล่นอัตโนมัติ
เล่นใหม่
23:06 last line (for ISTA), do we need \eta before 2A^T(Ax - y)?
Yes, you're correct -- the video has a typo. Thanks! Proximal gradient in general, for the composite problem: min: g(x) + h(x), is: prox_{\eta h}(x - \eta abla g(x)) (as in lectures 5.2 and 5.3).
![Neural networks [8.2] : Sparse coding - inference (ISTA algorithm)](/img/n.gif)
23:06 last line (for ISTA), do we need \eta before 2A^T(Ax - y)?
Yes, you're correct -- the video has a typo. Thanks! Proximal gradient in general, for the composite problem: min: g(x) + h(x), is: prox_{\eta h}(x - \eta
abla g(x)) (as in lectures 5.2 and 5.3).