by Andrew Ng (Stanford University & DeepLearning.AI)
This repository serves as a comprehensive archive of my journey through the Machine Learning Specialization. It includes hands-on programming assignments, theoretical notes, and implementation of core algorithms from scratch.
The specialization covers the modern machine learning pipeline, from fundamental statistical models to advanced neural network architectures.
- Supervised Learning: Linear/Logistic Regression, Regularization (L1/L2)
- Advanced Learning Algorithms: Neural Networks, Decision Trees, Random Forests, and XGBoost
- Unsupervised Learning: K-means Clustering, Anomaly Detection, and Recommender Systems (Collaborative Filtering)
- ML Strategy: Bias/Variance analysis, Error analysis, and Data Augmentation
├── 01-Supervised-Learning/ # Linear & Logistic Regression
├── 02-Advanced-Algorithms/ # Neural Networks & Decision Trees
├── 03-Unsupervised-Learning/ # Clustering, Recommenders, Reinforcement Learning
├── assignments/ # Completed Coursera Labs (.ipynb)
└── projects/ # Custom implementations and extra datasets
Throughout this course, I implemented several key algorithms both using frameworks and from scratch to understand the underlying math:
| Algorithm | Key Concepts | Tools Used |
|---|---|---|
| Linear Regression | Gradient Descent, Feature Scaling | NumPy, Matplotlib |
| Logistic Regression | Sigmoid Function, Binary Cross-Entropy | NumPy, Scikit-learn |
| Neural Networks | Backpropagation, Multiclass Classification | TensorFlow, Keras |
| Decision Trees | Information Gain, Entropy, Random Forests | Scikit-learn, XGBoost |
| Anomaly Detection | Gaussian Distribution | NumPy |
- Course 1: Supervised Machine Learning: Regression and Classification (In Progress)
- Course 2: Advanced Learning Algorithms
- Course 3: Unsupervised Learning, Recommenders, Reinforcement Learning
- Implemented gradient descent optimization from scratch, achieving convergence within 1000 iterations
- Built a neural network classifier with 95%+ accuracy on the MNIST dataset
- Developed an anomaly detection system using Gaussian distribution modeling
- Trained XGBoost models for structured data with hyperparameter tuning
Feel free to reach out if you'd like to discuss machine learning concepts or collaborate on projects!
- LinkedIn: Reema Hanim H
- Email: reemahanimh@gmail.com
"Artificial Intelligence is the new electricity." — Andrew Ng
Last Updated: January 2026