A-Z Linear Algebra & Calculus for AI & Data Science
Module 01: Vectors, Matrices & Linear Transformations
● Basic Properties: Notation, Terminologies, Sets, Subsets, Scalars, Vectors, and Matrices
● Vector Spaces & Basis (Subspaces, Linear Independence, Dimension)
● Introduction to Norms
● Column Space, Row Space, Null Space
● Matrix Operations: Addition, Subtraction, Multiplication
● Rank, Trace, Transpose Matrix
● Identity, Diagonal, Sparse & Dense Matrix
● Determinant, Orthogonal, Singular, Non-singular Matrix
● Inverse Matrix, Alternative to Matrix Division
● Covariance and Symmetric Matrices
● Positive-Negative Definite Matrix, Positive Semi-Definite
● Basic Logarithmic Properties
● Linear Transformations, Kernel & Range
● Change of Basis & Similarity Transformations
● Gram Matrix (Used in Kernel Methods)
● Eigenvalues & Eigenvectors
● Matrix Reduction: Cramer’s Rule, Gaussian Elimination
Module 02: Differential Calculus for Machine Learning
● Basics of Differentiation
● Differentiation Rules: Power, Constant, Exponential, Product, Quotient, Chain Rule
● 1st & 2nd Order Partial Derivatives
● Convex vs. Non-Convex Functions
● Gradient Descent Concepts for Optimization
● Hessian Matrix & Its Role in Optimization
Module 03: Integral Calculus for Machine Learning
● Indefinite Integral, Definite Integral
● Continuous Integral Form of MSE, MAE, RMSE
● Probability Density Function (PDF)
● Cumulative Distribution Function (CDF)
● Expectation (Mean), Variance
● Sigmoid Function
● Shannon Entropy
● KL Divergence
www.aiquest.org || https://youtube.com/studymart
Module 04: Regression Analysis & Correlation Analysis
● Correlation vs. Causation vs. Covariance
● Pearson Correlation / Types of correlations
● Covariance and Its Interpretation
● Simple & Multiple Linear Regression
● MSE, MAE, and RMSE
● R-Squared for Model Evaluation
● Distance Metrics: Euclidean Distance (L2 norm), Manhattan Distance (L1 norm),
Minkowski Distance, Mahalanobis Distance, Chebyshev Distance, Cosine Similarity,
Hamming Distance, and Jaccard Distance
Module 05: More About Regression in Machine Learning
● Polynomial Regression
● Real Problems & Solutions with High-Dimensional Data
● L1 (Lasso) Regularization
● L2 (Ridge) Regularization
● Elastic Net (L1 & L2)
Module 06: Matrix Factorization & Dimensionality Reduction
● Determinants & Rank of a Matrix
● Singular Value Decomposition (SVD)
● Low-Rank Approximation using SVD
● Moore-Penrose Pseudoinverse
● PCA for High-Dimensional Data
● t-SNE for Visualization
Module 07: Advanced Matrix Operations & Decompositions
● QR Decomposition & LU Decomposition
● Cholesky Decomposition for Positive Definite Matrices
● Adjoint & Pseudoinverse of a Matrix
● Norms: L1, L2, Frobenius, Spectral Norms
Module 08: Tensor Algebra for Deep Learning
● Tensors vs. Matrices
● Tensor Operations & Contractions
● Eigenvalues & Singular Values for Tensors
● Application of Tensor Algebra in Neural Networks
www.aiquest.org || https://youtube.com/studymart
Special Module: Probabilistic Models and Bayesian Methods
● Bayes’ Theorem & Conditional Probability
● Bayesian Inference: Prior, Likelihood, Posterior
● Probability Distribution for Discrete & Continuous Data
● Markov Chains & Monte Carlo Methods (MCMC)
● Hidden Markov Models (HMMs)
● Variational Inference
● Baysian Neural Network
● Baysian GANs
Instructor:
Rashedul Alam Shakil
MSc in Data Science at University of Erlangen (FAU)
Founder of Study Mart & aiQuest Intelligence
Automation Programmer at Siemens Energy
www.aiquest.org || https://youtube.com/studymart
