Applied Machine Learning
Robert Utterback
08/24/2022
1. Logistics
- Course website: https://robertutterback.github.io/courses/comp347/f22
2. Machine Learning vs. Coding
3. What is Machine Learning?
3.1.
3.2.
3.3.
3.4.
3.5.
3.6.
3.7.
4. Types of Machine Learning
4.1. Types of Machine Learning
- Supervised
- Unsupervised
- Reinforcement
4.2. Supervised Learning
\((x_i,y_i) \propto p(x,y)\) i.i.d
\(x_i \in \mathbb{R}^p\)
\(y_i \in \mathbb{R}\)
\(f(x_i) \approx y_i\)
4.3. Examples of Supervised Learning
- Testing for diabetes
- Classifying terrain of satellite image
- Automate manual labor
4.4. Unsupervised Learning
\(x_i \propto p(x)\) i.i.d
Learn about \(p\).
4.5. Reinforcement Learning
4.6. Reinforcement Learning
- Working with an environment, not a dataset
- Action influence the environment
- Can't look at all possible situation
- no separate data collection and learning
4.7. Other Kinds of Learning
- Semi-supervised
- Active Learning
- Forecasting
- …
5. Supervised Learning
5.1. Classification and Regression
Classification
- target \(y\) discrete
- Will you pass the class?
Regression
- Target \(y\) continuous
- How many points will you get on the final?
5.2. Generalization
Not only \(f(x_i) \approx y_i\),
also for new data: \(f(x) \approx y\)
5.3. Relationship to Statistics
Statistics
- model first
- inference emphasis
Machine Learning
- data first
- prediction emphasis
6. Overview of the course
- Infrastructure and basic tools
- Basics with end-to-end Case Studies
- Classification
- Training Models (optimizing)
- Support Vector Machines
- Decision Trees & Ensembles
- Dimensionality Reduction
- Unsupervised Learning
- Working with Text Data
- Neural Networks & Deep Learning