Eikosstatistics & data science

Machine Learning & Artificial Intelligence · Data Mining & Feature Engineering

Data mining · model evaluation, validation, and selection

Public introduction · full guide in Eikos

Introduction

What this guide covers

Introduction Introduction Advanced Data Mining · Statistical and Theoretical Framework

Overview

Topics covered

  • Introduction
  • Evaluation, Validation and Selection of Models
  • Fundamentals of Generalization and the Bias-Variance Dilemma
  • The Bias-Variance Decomposition in Regression
  • The Decomposition in Classification (Loss 0-1)
  • Hypothesis and Complexity Space: Vapnik-Chervonenkis Theory

Related concepts

Main connections

  • Variance & standard deviation
  • Overfitting & cross-validation
  • Bootstrap resampling
  • Cross-validation & train/test splits
  • Mean & expected value
  • Bayes' theorem & priors

Compatible notation

Mathematical sample in LaTeX

The expression is rendered accessibly, with its LaTeX source preserved below.

𝒟n={(x1,y1),(x2,y2),…,(xn,yn)}⊂(𝒳×𝒴)n
View LaTeX source\mathcal{D}_n = \{(x_1, y_1), (x_2, y_2), \dots, (x_n, y_n)\} \subset (\mathcal{X} \times \mathcal{Y})^n