DOMAIN 01 • COGNITIVE COMPUTING
ARTIFICIAL INTELLIGENCE PROGRAMS
Artificial Intelligence represents the frontier of modern computing, automating complex reasoning, perception, and decision-making. Our comprehensive AI curriculum empowers students with end-to-end expertise in cognitive algorithms, machine learning pipelines, deep neural architectures, computer vision, and modern language processing.
Foundational paradigms of intelligent systems, heuristic search algorithms, problem-solving techniques, knowledge representation, expert systems, and adversarial search.
- Heuristic & uninformed search algorithms (BFS, DFS, A*)
- Propositional & First-Order predicate logic
- Game playing algorithms (Minimax, Alpha-Beta Pruning)
- Probabilistic reasoning & Bayesian belief networks
Practical model training using Scikit-Learn, covering regression analysis, classification, decision trees, random forests, clustering techniques, and model evaluation metrics.
- Feature engineering, scaling & dimensional reduction (PCA)
- Supervised learning: SVM, KNN, Ensemble methods & XGBoost
- Unsupervised learning: K-Means, Hierarchical Clustering
- Cross-validation, hyperparameter tuning & deployment
Deep neural networks engineered with PyTorch and TensorFlow, focusing on backpropagation, convolutional layers, sequence models, optimizers, and loss regularization.
- Perceptrons, activation functions & gradient descent dynamics
- Deep feedforward networks & weight initialization strategies
- Vanishing / Exploding gradient mitigation & batch normalization
- Building custom neural architectures in PyTorch
Visual perception engineering with OpenCV and CNNs, covering spatial filtering, feature extraction, real-time YOLO object detection, image segmentation, and face tracking.
- Digital image fundamentals, color spaces & morphological filtering
- Edge detection (Canny, Sobel) & contour analysis with OpenCV
- Convolutional Neural Networks (ResNet, VGG, MobileNet)
- Real-time object tracking, bounding box regression & YOLOv8
Text intelligence covering tokenization, sentiment classification, named entity recognition, Transformer self-attention mechanisms, and fine-tuning modern Large Language Models.
- Text preprocessing with NLTK, spaCy & regular expressions
- Word2Vec, GloVe & semantic vector spaces
- Attention mechanisms, BERT, GPT & HuggingFace libraries
- Sentiment analysis, text summarization & chatbot engineering