Machine Learning Assignments

This repository contains a collection of assignments from various machine learning courses, covering topics ranging from basic linear regression to advanced deep reinforcement learning techniques.

Table of Contents

• Course 1
  • Linear Regression
  • Logistic Regression
  • Gradient Descent Algorithms
• Course 2
  • Handwritten Digit Recognition with TensorFlow
  • Bias and Variance Reduction Techniques
• Course 3
  • Unsupervised Learning Methods
  • Recommender Systems
  • Deep Reinforcement Learning

Course 1

Linear Regression

In this module, we explored the basics of linear regression, including the following:

• Understanding the linear regression model.
• Implementing linear regression using normal equations.
• Evaluating model performance using metrics such as Mean Squared Error (MSE) and R-squared.

Files:

• linear_regression.ipynb - Jupyter notebook implementing linear regression.

Logistic Regression

This module focused on logistic regression for binary classification tasks:

• Understanding the logistic function and its use in binary classification.
• Implementing logistic regression using gradient descent.
• Evaluating model performance using metrics such as accuracy, precision, recall, and F1-score.

Files:

• logistic_regression.ipynb - Jupyter notebook implementing logistic regression.

Gradient Descent Algorithms

This module covered various gradient descent algorithms:

• Understanding the gradient descent algorithm and its variants (Batch, Stochastic, Mini-batch).
• Implementing gradient descent for linear and logistic regression.
• Comparing the performance and convergence rates of different gradient descent algorithms.

Files:

• gradient_descent.ipynb - Jupyter notebook implementing various gradient descent algorithms.

Course 2

Handwritten Digit Recognition with TensorFlow

In this module, we applied TensorFlow to the problem of handwritten digit recognition:

• Understanding the basics of neural networks.
• Implementing a neural network using TensorFlow to recognize handwritten digits from the MNIST dataset.
• Evaluating model performance using accuracy and confusion matrix.

Files:

• mnist_tensorflow.ipynb - Jupyter notebook implementing handwritten digit recognition with TensorFlow.

Bias and Variance Reduction Techniques

This module explored techniques for reducing bias and variance:

• Understanding the bias-variance tradeoff.
• Implementing techniques such as cross-validation, regularization, and ensemble methods.
• Evaluating the impact of these techniques on model performance.

Files:

• bias_variance_reduction.ipynb - Jupyter notebook implementing bias and variance reduction techniques.

Course 3

Unsupervised Learning Methods

In this module, we explored various unsupervised learning methods:

• Understanding clustering algorithms (K-means, Hierarchical clustering).
• Implementing dimensionality reduction techniques (PCA, t-SNE).
• Evaluating the performance of unsupervised learning algorithms.

Files:

• unsupervised_learning.ipynb - Jupyter notebook implementing various unsupervised learning methods.

Recommender Systems

This module covered the basics of recommender systems:

• Understanding collaborative filtering and content-based filtering.
• Implementing a simple recommender system.
• Evaluating the performance of recommender systems using metrics such as RMSE and precision@k.

Files:

• recommender_systems.ipynb - Jupyter notebook implementing a recommender system.

Deep Reinforcement Learning

In this module, we explored deep reinforcement learning:

• Understanding the basics of reinforcement learning and the Bellman equation.
• Implementing deep Q-learning.
• Evaluating the performance of reinforcement learning algorithms.

Files:

• deep_reinforcement_learning.ipynb - Jupyter notebook implementing deep reinforcement learning.

License

This project is licensed under the MIT License - see the LICENSE file for details.