+ All Categories
Home > Documents > Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill...

Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill...

Date post: 11-Sep-2021
Category:
Upload: others
View: 1 times
Download: 0 times
Share this document with a friend
14
Mathematics for Machine Learning Special Mathematics Lecture Nagoya University, Fall 2020 https://www.henrikbachmann.com/mml_2020.html Lecture 11: Neural Networks I
Transcript
Page 1: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Mathematics for Machine LearningSpecial Mathematics LectureNagoya University, Fall 2020

https://www.henrikbachmann.com/mml_2020.html

Lecture 11: Neural Networks I

Page 2: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Machine learning overview

Remaining plan:• Introduce (simple) Neural Networks (Today)• Understand how to train them (Next time)• Consider more complicated NN.• Use finished implementations, e.g. Tensorflow.

Page 3: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Nice to watch

https://www.3blue1brown.com/neural-networks

https://nnfs.io/

3blue1brown

Neural Networks from Scratch

Page 4: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Neural Networks

Page 5: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Recall: Logistic regression

Logistic regression

Hypothesis:

Example: Binary classifier (Pass an exam Yes/No, Spam email Yes/no)

We learned the correct parameters by maximizing the log-likelihood (by using gradient ascent)

Or mizimizing the negative of the log-likelihood (= cost function) (by using gradient descent)

Page 6: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Rewriting logistic regression as a neural network

Page 7: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Motivation: Image classification

Goal: Check if a picture contains a cat

Page 8: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Motivation: Image classification

Goal: Check if a picture contains a cat or dog

Page 9: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Motivation: Image classification. More layers

Goal: Check if a picture contains a cat or dog

Page 10: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

(rough) Notation

Page 11: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Content of a layer

Page 12: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Content of a layer

Page 13: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Example of activation functionsThere are several common activiation functions.

https://en.wikipedia.org/wiki/Activation_function

Question: Why not use the identity map as an activation function?

Page 14: Mathematics for Machine Learning - Henrik Bachmann · educba.com Supervised Learning Game Al Skill Acquisition Identity Fraud Feature Detection Elicitation Machine Learning Recommender

Example of NNLet us consider a NN with 4 layers (2 hidden):


Recommended