ML Learning Vault

Personal library of machine learning, deep learning, and reinforcement learning tutorials, implementations, and experiments collected over 3+ years.

📊 Stats

• 60+ projects across multiple ML domains
• Deep Learning: PyTorch, Neural Networks from scratch
• Reinforcement Learning: DQN, PPO, TD3, Rainbow, game environments
• LLM Agents: CrewAI, AutoGen, multi-agent systems
• Neural ODEs: torchdyn, differential equations for ML
• Hyperparameter Optimization: Optuna, custom algorithms

🗂️ Repository Structure

ml-learning-vault/
├── deep_learning/              # Neural networks, PyTorch fundamentals
├── reinforcement_learning/     # RL algorithms and game environments
│   ├── basics/                # DQN, DDPG, PPO, TD3
│   ├── advanced/              # Rainbow DQN, advanced techniques
│   └── environments/          # Game AIs (Snake, DOOM, Mario, etc.)
├── llm_agents/                # CrewAI, AutoGen, agent frameworks
├── neural_odes/               # Neural Ordinary Differential Equations
├── hyperparameter_optimization/ # Optuna, optimization techniques
├── libraries_tutorials/       # skrl, RLlib, Polars
└── programming_fundamentals/  # Docker, Git, Python basics

🔍 Quick Search

Use the search script to find projects:

# General search
./search.sh "DQN"

# Search by tags
./search.sh -t "RL,PyTorch"
./search.sh -t "Games"

# Search by category
./search.sh -c "RL"
./search.sh -c "LLM"

# Search by type
./search.sh -type "Tutorial"
./search.sh -type "Project"

# Search by status
./search.sh -s "Complete"

📖 Full Catalog

See CATALOG.md for a complete table of all 60+ projects with:

• Project descriptions
• Technology tags
• Status tracking
• Source attribution

🎯 Featured Topics

Reinforcement Learning

• Algorithms: DQN (PyTorch & TensorFlow), DDPG, PPO, TD3, Rainbow DQN
• Frameworks: skrl, RLlib, custom implementations
• Environments: Snake AI, Connect Four, Chrome Dino, DOOM, Street Fighter, Super Mario

LLM & Agents

• CrewAI: Multiple courses (Maya Akim, CodeWithBrandon, master course)
• AutoGen: Multi-agent systems
• Custom: AI Data Assistant

Neural ODEs

• Basic implementations
• torchdyn framework tutorials
• Continuous depth models

Hyperparameter Optimization

• Optuna (Bayesian optimization)
• Custom algorithms
• Grid search & random search alternatives

📝 Usage

Each project contains:

• requirements.txt - Python dependencies (with Python version commented)
• Source code organized in logical structure
• Documentation where applicable

To use a project:

cd reinforcement_learning/basics/dqn_pytorch
pip install -r requirements.txt
python main.py  # or follow project-specific instructions

🏷️ Tags Legend

• RL: Reinforcement Learning
• DL: Deep Learning
• LLM: Large Language Models
• PyTorch: PyTorch framework
• TensorFlow: TensorFlow framework
• Games: Game AI environments
• Agents: Multi-agent systems
• HyperOpt: Hyperparameter optimization
• NODE: Neural ODEs

📊 Organization Philosophy

This vault serves as both:

1. Learning archive: Tracking ML journey from basics to advanced topics
2. Reference library: Quick access to implementations and techniques
3. Code templates: Reusable patterns for future projects

Projects are organized by domain rather than chronology, making it easy to find related work across different learning sources.

---

Last Updated: February 2026 Total Projects: 60+ Active Domains: 7