roboticcam/machine-learning-notes
Machine Learning Notes
This project is a machine learning study guide and technical knowledge base. It serves as a version-controlled repository of mathematical formulas and algorithmic explanations, providing instructional material and reference notes for the study of artificial intelligence.
The content is structured as a markdown-based knowledge base that pairs theoretical mathematical explanations directly with code implementations. This approach demonstrates model mechanics in practice across several specialized domains, including deep learning research, probabilistic graphical modeling, and reinforcement learning theory.
The curriculum covers a broad technical surface, including foundational machine learning mathematics, 3D computer vision geometry, and generative AI architectures. It also includes detailed material on probabilistic inference, optimization methods, and natural language processing.
Features
- Machine Learning Education - Serves as a comprehensive educational resource for the mathematical and theoretical foundations of machine learning.
- Deep Learning Research - Explores advanced neural network architectures and generative models through a research-oriented lens.
- Machine Learning Foundations - Explores the foundational mathematics, probability, and statistics that form the theoretical basis of machine learning.
- Machine Learning Guides - Collects technical notes and code implementations covering the mathematics and architectures of machine learning.
- Probabilistic Graphical Models - Provides detailed technical documentation and implementations for Bayesian inference, Monte Carlo methods, and state space models.
- Markdown-Based Knowledge Bases - Stores theoretical explanations and mathematical formulas in version-controlled markdown files for portability.
- Annotated Code Implementations - Pairs theoretical mathematical explanations directly with executable code snippets to demonstrate model mechanics.
- Deep Learning Fundamentals - Covers core neural network concepts, including convolutional neural networks and loss functions, with practical implementation.
- Machine Learning Mathematics - Provides foundational mathematics covering model evaluation, decision trees, probability, and regression for machine learning.
- AI & Machine Learning Education - Provides foundational educational content covering artificial intelligence and machine learning algorithms and their practical implementations.
- Deep Learning Reference Implementations - Provides detailed technical explanations of neural networks paired with concrete code implementations.
- Deep Learning Architectures - Reviews technical specifications and structural compositions of convolutional and graph neural networks.
- Generative AI Learning Resources - Offers instructional notes and specialized research on generative models, transformers, and variational autoencoders.
- Generative AI Architectures - Examines transformer structures and attention mechanisms by implementing key-value caching and core model mechanics.
- Study Guides - Provides instructional material on 3D reconstruction and epipolar geometry for depth estimation.
- Deep Learning Optimization - Evaluates research on implicit bias and duality to improve convergence for stochastic gradient descent.
- Gradient Descent Algorithms - Analyzes optimization algorithms including Lagrangian duality, KKT conditions, and conjugate gradient descent.
- Probabilistic Models - Models complex data using expectation maximization, Markov Chain Monte Carlo, and variational inference.
- Transformer Architecture Implementation - Implements sequence-to-sequence architectures including rotary positional embeddings and multi-head latent attention.
- Bayesian Inference - Estimates unknown variables by applying Monte Carlo methods, particle filtering, and Bayesian non-parametrics.
- Geometry and Vision - Provides resources for reconstructing 3D spaces using camera models and epipolar geometry.
- Curricula - Provides instructional material on 3D reconstruction, camera models, and epipolar geometry.
- Learning Paths - Structures theoretical learning paths starting from foundational linear algebra and probability before moving to complex architectures.
- Mathematics Study Guides - Provides lecture notes and formula derivations for intermediate ML math, including expectation maximization and variational inference.
- Natural Language Processing Resources - Provides educational content and academic breakdowns of natural language processing, specifically focusing on word embeddings and attention mechanisms.
- Reinforcement Learning Theory - Explores the mathematical and theoretical foundations of reinforcement learning, including Markov Decision Processes and policy gradients.
- Curriculum Decomposition - Breaks complex subjects like 3D vision and probabilistic inference into discrete notes for incremental learning.
- Topic-Based Resource Organization - Organizes educational content into a nested folder structure based on machine learning domains and mathematical prerequisites.
- Epipolar Geometry - Implements depth and pose estimation using camera models and epipolar geometry.
- Monte Carlo Sampling - Executes probabilistic simulations through inverse CDF sampling, importance sampling, and particle filters.
- Technical Knowledge Maps - Connects diverse fields like 3D vision and probabilistic inference through a unified system of interlinked technical notes.
