Learning

This section contains four pieces: machine learning, statistical learning, statistical inference, and data science. Although each of these subjects are related, they are nonetheless distinct in important facets.

Machine learning is a technology of learning algorithms whose construction requires sample data. These learning algorithms can be thought of as models.

Statistical learning is a technology to interpret, assess, and improve the performance of learning algorithms. In particular, it should be “model independent.”

Statistical inference is a technology to give statistical evidence of one hypothesis over another.

Data science combines these technologies in order to influence decision making.

Contents

• Machine Learning
  • On the “Assessment” of Machine Learning Algorithms
  • Introduction to Topological Data Analysis
  • Machine Learning Researchers Suck at Math
  • The Zen of TDA
  • Genetic Algorithm Notes
• Statistical Learning
  • Hypotheses, Models and Loss Functions
  • Assessing Learners
  • From Samples to Probability Distributions
• Statistical Inference
  • The Permutation Test
  • Elements of Survival Analysis
  • The Theory Maximum Likelihood Estimators
  • An Introduction to MLE’s
  • The Statement of Fundamental Theorem of MLEs
  • Asymptotic Computations of MLEs
  • Relative Entropy of Gaussian Distributions
• Data Science
  • Constituent Ingredients of Data Science
  • Types of Data Scientists
  • How I Structure My Jupyter Notebooks
  • Resources