Reference

Quick cheat-sheets to keep open while you read papers or write code — symbols, algebra and log rules, derivatives, matrix and vector operations, NumPy calls, shape reasoning, and the equations that recur across ML.

• Mathematical Symbols

  The operators, relations, and set symbols that recur in ML papers.

• Greek Letters

  How each Greek letter is conventionally used in machine learning.

• Algebra Rules

  Exponents, factoring, and the identities you rearrange most often.

• Logarithm Rules

  Product, quotient, power, and change-of-base rules for logs.

• Derivative Rules

  Power, product, quotient, and chain rules with common derivatives.

• Matrix Operations

  Shapes, products, transpose, inverse, trace, and determinant.

• Vector Operations

  Dot product, norms, distance, angle, and cosine similarity.

• NumPy Operations

  The NumPy calls that implement each math operation, with shapes.

• Shape Reasoning

  Rules for predicting the shape of every operation before you run it.

• Common ML Equations

  MSE, linear models, cosine similarity, normalization, regularization.