Week 7, Convolutional Neural Network (CNN) What

Week 7, Convolutional Neural Network (CNN) What
1 / 1
Week 7, Convolutional Neural Network (CNN) What - slide 1 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 2 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 3 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 4 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 5 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 6 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 7 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 8 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 9 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 10 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 11 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 12 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 13 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 14 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 15 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 16 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 17 of 18 Week 7, Convolutional Neural Network (CNN) What - slide 18 of 18
Week 7, Convolutional Neural Network (CNN) What deep network can do? Computer Vision Classification Detection Segmentation Caption generation Large language models Protein folding ML-based force field, etc MLP and inductive bias In

Related Topics

Download this presentation From Below

"Week 7, Convolutional Neural Network (CNN) What" is the property of its rightful owner. Permission is granted to download and print the materials on this website for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.

Presentation Transcript

01
Week 7, Convolutional Neural Network (CNN)<br>
02
What deep network can do? Computer Vision
Classification
Detection
Segmentation
Caption generation Large language models

Protein folding

ML-based force field, etc<br>
03
MLP and inductive bias In principle, with enough width and depth, the multi-layer perceptrons (MLP) can do everything

However, in practice, it is inefficient due to a larger number of parameters needed for training and inference

Thus, a strong “inductive bias” is needed for computer vision/classification. Translational equi-variance is helpful<br>