CIS 365: Applied Artificial Intelligence

This repository showcases the core concepts and assignments completed as part of CIS 365: Applied Artificial Intelligence. The course focuses on the science and techniques behind designing intelligent systems capable of performing tasks that typically require human intelligence.

---

Overview

Artificial Intelligence (AI) is the science of designing machines that can perceive, learn, reason, and solve problems. This course explored foundational and advanced AI techniques, with applications spanning domains like machine learning, computer vision, and natural language processing.

Key Topics Covered:

• Search Algorithms
• Probability and Uncertainty
• Information Theory
• Machine Learning Models
• Deep Learning Architectures
• Generative Networks and Natural Language Processing

---

Table of Contents

Concepts

1. Uninformed Search
   Techniques like breadth-first, depth-first, and depth-limited search, which explore state spaces without additional domain knowledge.

2. Informed Search
   Heuristic-based algorithms like A*, guiding searches using estimated costs to improve efficiency.

3. Uncertainty in AI
   Handling incomplete information using probability theory, fuzzy logic, and other methods for decision-making.

4. Probability
   Essential for modeling uncertain scenarios, covering distributions, conditional probability, and Bayes’ theorem.

5. Information Theory
   Concepts like entropy and information gain, used in decision tree learning and data compression.

6. K-Nearest Neighbors (KNN)
   An instance-based learning algorithm for classification and regression based on proximity to known data.

7. Perceptron
   A foundational artificial neuron model, forming the basis of neural networks for binary classification tasks.

8. Computer Vision
   Enabling machines to process and interpret visual data, covering image histograms, smoothing, and feature extraction.

9. Convolutional Neural Networks (CNN)
   Deep learning models designed for tasks like image recognition, featuring convolutional and pooling layers.

10. Autoencoder
    Neural networks for unsupervised learning, useful in data compression and noise reduction tasks.

11. Dimensionality Reduction
    Techniques like PCA for reducing dataset features, improving visualization and computational efficiency.

12. Generative Networks and Natural Language Processing (NLP)
    Advanced methods for creating realistic data and processing human language, enabling AI systems to generate text, images, or audio.

---

Assignments

1. Object-Oriented Game
   Explored AI-driven gameplay mechanics in a simple object-oriented game design.

2. BFS and DFS
   Implemented breadth-first and depth-first search for navigating state spaces.

3. A Search*
   Developed heuristic-based search algorithms to solve optimization problems.

4. Bayesian Classifier
   Applied Bayes’ theorem for probabilistic classification tasks.

5. Bayes Formula
   Practiced conditional probability and its applications in AI models.

6. Information Gain
   Quantified decision tree splits using entropy and information theory principles.

7. Confusion Matrix
   Evaluated classification models using metrics like accuracy, precision, and recall.

8. K-Nearest Neighbors (KNN)
   Built a KNN model for classification and regression tasks.

9. Perceptron Learning
   Implemented a single-layer perceptron for binary classification.

10. Computer Vision
    Explored image processing techniques and feature extraction.

11. Convolutional Neural Networks (CNN)
    Built and trained CNN models for image recognition.

12. Autoencoder
    Designed neural networks for data compression and reconstruction tasks.

13. Dimensionality Reduction (PCA)
    Implemented Principal Component Analysis to reduce dataset complexity.

14. Generative Adversarial Networks (GAN)
    Experimented with generative models to create realistic synthetic data.

15. Natural Language Processing (NLP)
    Processed and analyzed textual data using NLP techniques.

---

Repository Organization

The repository is divided into two main sections:

• lectures/: Contains PDF lecture slides, covering theoretical foundations and key AI topics.
• assignments/: Contains project-specific code and resources for all 15 assignments.

---

Lessons Learned

This course provided a comprehensive foundation in Artificial Intelligence, emphasizing:

• Search algorithms for problem-solving.
• Probabilistic models and their applications in uncertain environments.
• Machine learning and neural network architectures, including CNNs and Autoencoders.
• Practical experience in computer vision, generative models, and natural language processing.

Through these assignments, I gained hands-on experience in implementing AI models, evaluating their performance, and exploring real-world applications of intelligent systems.