Opening Machine Learning: Neural Networks and Decision Trees
Machine Learning: Neural Networks and Decision Trees
- Length
- 1 h 25 min
- Level
- Undergraduate
- Recorded
- UC Berkeley2018
Chapter 1 of 12: Before the lecture
Builds decision trees by information gain, then prunes splits that fail a significance test against random chance.
Decision trees and neural networks are two answers to the same problem of fitting training data while generalizing to unseen examples. Abbeel opens session 23 of Berkeley's CS 188 with AlexNet, convolutional layers and parameter sharing, then surveys captioning, visual question answering, speech recognition and machine translation. Inductive learning is then framed as approximating a target function within a hypothesis space, with Occam's razor and the bias-variance tradeoff. Decision trees follow: their doubly exponential hypothesis space, the recursive learning algorithm, and attribute selection by entropy and information gain. Fully grown trees overfit, and pruning statistically insignificant splits serves as regularization.
Decision trees and neural networks are two answers to the same problem of fitting training data while generalizing to unseen examples. Abbeel opens session 23 of Berkeley's CS 188 with AlexNet, convolutional layers and parameter sharing, then surveys captioning, visual question answering, speech recognition and machine translation. Inductive learning is then framed as approximating a target function within a hypothesis space, with Occam's razor and the bias-variance tradeoff. Decision trees follow: their doubly exponential hypothesis space, the recursive learning algorithm, and attribute selection by entropy and information gain. Fully grown trees overfit, and pruning statistically insignificant splits serves as regularization.
Histograms of Oriented Gradients for Human Detection
Karpathy & Fei-Fei, 2015; Donahue et al., 2015; Xu et al., 2015
Visual QA Challenge paper
This lecture
Lecture 23 of 25 of Introduction to Artificial Intelligence.
Choose a node to see how it connects to this lecture. What comes before is on the leftat the top. What comes after is on the rightat the bottom. The ring is what it teaches. People and places are above.
Press a chosen node again to open it.
Your comment is kept while you sign up.Sign in