Machine Learning for Engineers: Using data to solve problems for physical systems

Part I Fundamentals

1.0  Introduction

1.1.   Where machine learning can help engineers

1.3.   Machine learning to correct idealized models

2.      The Landscape of machine learning

2.1.   Supervised learning

2.1.1.      Regression

2.1.2.      Classification

2.1.3.      Time series

2.1.4.      Reinforcement

2.3.   Optimization

2.4.   Bayesian statistics

3.      Linear Models

3.1.   Linear regression

3.2.   Logistic regression

3.3.   Regularized regression

3.4.   Case Study: Determining physical laws using regularized regression

4.      Tree-Based Models

4.1.   Decision Trees

4.3.   BART

4.4.   Case Study: Modeling an experiment using random forest models

5.      Clustering data

5.1.   Singular value decomposition

5.2.   Case Study: SVD to standardize several time series

5.3.   K-means

5.4.   K-nearest neighbors

5.5.   t-SNE

5.6.   Case Study: The reflectance spectrum of different foliage

Part II Deep Neural Networks

6.      Feed-Forward Neural Networks

6.2.   Dropout

6.3.   Backpropagation

6.5.   Regression

6.6.   Classification

7.      Convolutional Neural Networks

7.1.   Convolutions

7.2.   Pooling

7.3.   Residual networks

7.4.   Case Study: Finding volcanoes on Venus

8.      Recurrent neural networks for time series data

8.1.   Basic Recurrent neural networks

8.3.   Attention networks

8.4.   Case Study: Predicting future system performance

9.      Unsupervised Learning with Neural Networks

9.1.   Auto-encoders

9.3.   Case study: Optimization using Inverse models

10.  Reinforcement learning

10.1.                    Case study: controlling a mechanical gantry

11.  Transfer learning

Part IV Appendices

A.      SciKit-Learn

B.      Tensorflow