Lecture 11
Found 3 free book(s)
Lecture 11: Detection and Segmentation - Stanford University
cs231n.stanford.eduLecture 11 - 24 May 10, 2017 Semantic Segmentation Idea: Fully Convolutional Input: 3 x H x W Predictions: H x W Design network as a bunch of convolutional layers, with downsampling and upsampling inside the network! High-res: D 1 x H/2 x W/2 High-res: D 1 x H/2 x W/2 Med-res: D 2 x H/4 x W/4 Med-res: D 2 x H/4 x W/4 Low-res: D 3 x H/4 x W/4
Lecture 11: Generative Models - Stanford University
cs231n.stanford.eduLecture 11 - May 9, 2019 Unsupervised Learning Data: x Just data, no labels! Goal: Learn some underlying hidden structure of the data Examples: Clustering, dimensionality reduction, feature learning, density estimation, etc. Holy grail: Solve unsupervised learning => understand structure of visual world 15 Supervised vs Unsupervised Learning
Lecture 11 Attention and Transformers - Stanford University
cs231n.stanford.eduLecture 11 - May 03, 2022 x 1 we are eating x 2 x 3 h 1 h 2 h 3 s 0 bread x 4 h 4 e 11 e 12 e 13 e 14 softmax a 11 a 12 a 13 14 From final hidden state: Initial decoder state s 0 Normalize alignment scores to get attention weights 0 < a t,i < 1 ∑ i a t,i = 1 Bahdanau et al, “Neural machine translation by jointly learning to align and ...