The PAMCET Learning Team is the institutional curriculum and content group behind course publishing on this platform. PAMCET (Professional and Academic Management Centre for Education and Training) is operated by Digitpen Hub Ltd, a company registered in Lagos, Nigeria. Courses credited to the PAMCET Learning Team are structured, organized, and published centrally by the platform rather than authored by a single named individual instructor — each course's syllabus, learning outcomes, and content are reviewed before publication under this institutional account. For questions about a specific course or to report a content issue, please use the platform's Help & Support or Contact Us pages.
Bachelor of Science (B.Sc.) in Artificial Intelligence: Algorithmic Foundations, Autonomous Agents & Machine Intelligence
A comprehensive 4-year undergraduate degree curriculum aligned with the National Universities Commission (NUC) Core Curriculum and Minimum Academic Standards (CCMAS) for Artificial Intelligence in Nigerian universities.
Course Overview
A comprehensive 4-year undergraduate degree curriculum aligned with the National Universities Commission (NUC) Core Curriculum and Minimum Academic Standards (CCMAS) for Artificial Intelligence in Nigerian universities. Curriculum Structure & Core Modules: • 1. Mathematical...
Degree Programme Curriculum Context
This course is a curriculum unit of the Bachelor of Science (B.Sc.) in Artificial Intelligence degree framework in Nigeria.
Earn Your Certificate
Showcase your skills with a certificate! Complete the course, submit your project, and earn your certificate. Here's a sample of what you'll receive to show off your accomplishments.

Course Curriculum
Browse complete theoretical foundations, worked calculations, laboratory protocols, and board assessment questions under each topic.
Vector Spaces, Matrix Decompositions (SVD & Eigendecomposition) & Tensor Operations in High Dimensions
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Vector Spaces, Matrix Decompositions & Tensor Operations in High Dimensions Module 1, Lecture 1: Disciplinary Foundations of High-Dimensional Linear Algebra and Spectral Geometry for Machine Intelligence. University Hub > Artificial Intelligence > Vector Spaces & Decompositions 1. Formal Vector Spaces and...
Multivariate Gradient Derivations, Jacobian & Hessian Matrices, and Constrained Optimization with KKT Conditions
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Applied Practice & Calculations Multivariate Gradient Derivations, Jacobian & Hessian Operators, and KKT Optimality Module 1, Lecture 2: Differential Calculus on Vector Spaces, Second-Order Curvature, and Constrained Convex Optimization. University Hub > Artificial Intelligence > Multivariate Gradients & KKT 1. Vector and Matrix D...
Summative Mathematical Problem Set: Projection Matrices, Spectral Radius & Convergence Proofs for Gradient Dynamics
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Mathematical Problem Set: Spectral Convergence & Optimization Bounds Module 1, Lecture 3: Rigorous Mathematical Derivations of Projection Properties, Spectral Radii, and Gradient Descent Convergence Rates. University Hub > Artificial Intelligence > Mathematical Problem Set & Proofs Problem...
Measure-Theoretic Probability, Maximum Likelihood Estimation (MLE) & Bayesian Inference Principles
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Measure-Theoretic Probability, Maximum Likelihood & Bayesian Inference Module 2, Lecture 1: Formal Probability Spaces, Likelihood Theory, Fisher Information, and Bayesian Parametric Updating. University Hub > Artificial Intelligence > Statistical Foundations & MLE 1. Formal Measure-Theoretic Probability F...
Directed and Undirected Graphical Models: D-Separation, Exact Inference with Variable Elimination & Belief Propagation
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Applied Practice & Calculations Directed & Undirected Graphical Models, D-Separation & Exact Inference Module 2, Lecture 2: Bayesian Networks, Markov Random Fields, D-Separation Topologies, and the Variable Elimination Algorithm. University Hub > Artificial Intelligence > Probabilistic Graphical Models 1. Directed Acyclic Graphs (...
Summative Problem Set: Factor Graphs, Hidden Markov Models & Expectation-Maximization (EM) Algorithm Derivation
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Statistical Problem Set: Factor Graphs & EM Convergence Module 2, Lecture 3: Rigorous Derivations of Hidden Markov Model Message Passing and the Expectation-Maximization Theorem. University Hub > Artificial Intelligence > HMM & EM Derivations Problem 1: Exact Forward-Backward Recursion in H...
Formal State-Space Formulations, Heuristic Admissibility & Consistency Theorems
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Formal State-Space Formulations & Heuristic Optimality Theorems Module 3, Lecture 1: State-Space Formalisms, Admissibility, Consistency, and A* Optimality Guarantees. University Hub > Artificial Intelligence > State-Space Search Foundations 1. Formal Mathematical Formulation of Deterministic Search Problems C...
Advanced Search Mechanics: IDA*, Bidirectional Search, Constraint Satisfaction (AC-3) & STRIPS Planning
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Applied Practice & Calculations Advanced Search Mechanics: IDA*, Bidirectional Search, CSPs & Planning Module 3, Lecture 2: Memory-Bounded Heuristic Search, Constraint Propagation via AC-3, and STRIPS Action Schema Formalisms. University Hub > Artificial Intelligence > Advanced Search & Planning 1. Memory-Bounded Search: Iterative...
Summative Search Problem Set: Heuristic Domination Proofs, Game Trees & Alpha-Beta Pruning Bounds
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Search Problem Set: Heuristic Domination & Alpha-Beta Bounds Module 3, Lecture 3: Formal Proof of Heuristic Domination Subsets and Asymptotic Complexity Derivation of Optimal Alpha-Beta Pruning. University Hub > Artificial Intelligence > Domination Proofs & Alpha-Beta Problem 1: Heuristic D...
Statistical Learning Theory: Empirical Risk Minimization, PAC Learnability & VC Dimension Bounds
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Statistical Learning Theory: PAC Learnability & VC Dimension Module 4, Lecture 1: Empirical Risk Minimization, Agnostic PAC Bounds, Shattering Coefficients, and Generalization Guarantees. University Hub > Artificial Intelligence > Statistical Learning Theory 1. Formal Statistical Framework and Empirical Risk...
Convex Optimization in Support Vector Machines: Dual Formulations, Mercer's Theorem & Reproducing Kernel Hilbert Spaces (RKHS)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Mathematical Derivations & Convex Optimization Convex Optimization in SVMs: Primal, Dual & the Kernel Trick Module 4, Lecture 2: Hyperplane Geometry, Wolfe Dual Formulation, Karush-Kuhn-Tucker Conditions, and Mercer Kernels in RKHS. University Hub > Artificial Intelligence > Support Vector Machines & Kernels 1. Hyperplane Geometry...
Summative Learning Theory Problem Set: Bias-Variance Decomposition, Rademacher Complexity & Boosting Convergence
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Learning Theory Problem Set: Proofs & Dual Solves Module 4, Lecture 3: Algebraic Bias-Variance Derivation, VC Dimension Shattering Proofs, and Explicit SVM Quadratic Optimization. University Hub > Artificial Intelligence > Learning Theory Problem Set Problem 1: Algebraic Bias-Variance-Noise Dec...
Universal Approximation Theorems, Computation Graphs & Reverse-Mode Automatic Differentiation
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Universal Approximation Theorems & Reverse-Mode Autodiff Module 5, Lecture 1: Functional Density in Banach Spaces, Depth-Width Trade-offs, and Vector-Jacobian Product Mechanics. University Hub > Artificial Intelligence > Neural Approximations & Autodiff 1. The Universal Approximation Theorem: Density in C...
Loss Surface Topography: Vanishing/Exploding Gradients, Batch Normalization Mechanics & Adaptive Optimizers (Adam, RMSprop)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Mathematical Derivations & Optimization Dynamics Loss Surface Topography, Normalization & ResNet Dynamics Module 5, Lecture 2: Hessian Curvature, Gradient Vanishing Multipliers, Batch Normalization Gradients, and Skip Connection Highways. University Hub > Artificial Intelligence > Optimization Dynamics & Architectures 1. Loss Surf...
Summative Architecture Problem Set: Weight Initialization Proofs (He/Xavier), Residual Connections & Gradient Flow
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Architecture Problem Set: Proofs & Numerical Traces Module 5, Lecture 3: Xavier and He Variance Derivations, Convolutional Receptive Field Calculus, and Hand-Computed Backpropagation. University Hub > Artificial Intelligence > Architecture Problem Set Problem 1: Variance Propagation and Xavier...
Distributional Semantics, Recurrent Information Flow & The Information Bottleneck Problem
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Distributional Semantics, RNN Flow & LSTM Cell Mechanics Module 6, Lecture 1: Skip-Gram Objective Derivations, BPTT Vanishing Gradients, and Constant Error Carousels. University Hub > Artificial Intelligence > Sequence Modeling & LSTMs 1. Distributional Semantics and Skip-Gram Negative Sampling (SGNS) Nat...
Scaled Dot-Product Attention: Multi-Head Projection, Positional Encodings & Computational Complexity Analysis
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Mathematical Derivations & Transformer Architecture Scaled Dot-Product Attention, RoPE & KV Caching Module 6, Lecture 2: Attention Normalization Proofs, Multi-Head Projections, Rotary Relative Embeddings, and Decoding Memory Footprints. University Hub > Artificial Intelligence > Attention & Transformers 1. Scaled Dot-Product Atten...
Summative Transformer Problem Set: Self-Attention Gradient Derivations, KV Caching & Scaling Law Mechanics
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Transformer Problem Set: Proofs & Scaling Laws Module 6, Lecture 3: Softmax Jacobian Derivation, Chinchilla Compute Optimality Proofs, and KV Cache Footprints. University Hub > Artificial Intelligence > Transformer Proofs & Scaling Problem 1: Analytical Softmax and Self-Attention Jacobian D...
Markov Decision Processes (MDPs), Bellman Optimality Equations & Dynamic Programming Formulations
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Markov Decision Processes & Bellman Optimality Theorems Module 7, Lecture 1: Bellman Expectation and Optimality Equations, Contraction Mapping Proofs, and Dynamic Programming. University Hub > Artificial Intelligence > Reinforcement Learning Foundations 1. Mathematical Formulation of Markov Decision Processes...
Model-Free Control: Temporal Difference Learning, Q-Learning Convergence & Policy Gradient Theorem Derivation
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Mathematical Derivations & Control Theory Model-Free Control, Q-Learning & Policy Gradient Theorem Module 7, Lecture 2: Temporal Difference Errors, Off-Policy Q-Learning Convergence, and the Analytical Policy Gradient Theorem. University Hub > Artificial Intelligence > Model-Free RL & Policy Gradients 1. Temporal Difference Learni...
Summative RL Problem Set: Advantage Actor-Critic (A2C), Generalized Advantage Estimation & Proximal Policy Optimization
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative RL Problem Set: Policy Proofs & PPO Dynamics Module 7, Lecture 3: Trajectory Policy Gradient Proof, Baseline Variance Reduction, GAE Derivations, and PPO Clipping Bounds. University Hub > Artificial Intelligence > Policy Proofs & PPO Problem 1: First-Principles Proof of the Policy Gradient...
Formal Mathematical Formulations of Fairness: Demographic Parity, Equalized Odds & Calibration
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Core Theory & Foundations Algorithmic Fairness, Differential Privacy & Recourse Module 8, Lecture 1: Mathematical Fairness Criteria, Impossibility Theorems, (\epsilon, \delta)-Differential Privacy, and Counterfactual Recourse. University Hub > Artificial Intelligence > Fairness & Privacy Foundations 1. Formal Mathematical Criteria...
Robust Machine Learning: Distribution Shift, Uncertainty Quantification (Conformal Prediction) & Differential Privacy
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Mathematical Derivations & Robustness Theory Robust Machine Learning: Shift, Conformal Sets & Defenses Module 8, Lecture 2: Covariate Shift Importance Weighting, Distribution-Free Conformal Prediction, and Minimax Robust Optimization. University Hub > Artificial Intelligence > Robustness & Conformal Prediction 1. Distribution Shif...
Summative Safety Problem Set: Reward Gaming, Value Alignment Formalisms & Regulatory Governance Frameworks
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Summative Problem Set & Proofs Summative Safety Problem Set: Fairness Impossibility & Alignment Module 8, Lecture 3: Algebraic Proof of Fairness Trade-offs, DP Laplace Budgeting, and Bradley-Terry Preference Derivations. University Hub > Artificial Intelligence > Safety Proofs & Alignment Problem 1: Algebraic Proof of the Impossib...
Graph Theory Fundamentals, Adjacency Matrices & Spectral Graph Properties
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 9.1: Discrete Mathematics Graph Theory Fundamentals, Adjacency Matrices & Spectral Graph Properties Rigorous algebraic and topological foundations of discrete relational structures in machine intelligence, spectral graph theory, and Laplacian partitioning. Academic Learning Outcomes Formulate discrete relational systems using formal graph-th...
Propositional and First-Order Predicate Logic, Inference Rules & Resolution Proofs
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 9.2: Discrete Mathematics Propositional and First-Order Predicate Logic, Inference Rules & Resolution Proofs Formal logical reasoning frameworks, syntax, model-theoretic semantics, Herbrand interpretations, Skolemization, and Robinson's resolution refutation principle in knowledge-based systems. Academic Learning Outcomes Formalize natur...
Summative Discrete Math Lab: Graph Traversal Algorithms & Automated Theorem Proving
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 9.3: Practical Lab Studio Summative Discrete Math Lab: Graph Traversal Algorithms & Automated Theorem Proving Hands-on engineering laboratory synthesizing algorithmic graph traversal (DFS/BFS/Topological sorting), shortest path networks, and building a full Davis-Putnam-Logemann-Loveland (DPLL) Boolean satisfiability solver. Academic Lea...
Asymptotic Analysis (Big-O, Omega, Theta), Recurrence Relations & Master Theorem
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 10.1: Algorithmic Complexity Asymptotic Analysis (Big-O, Omega, Theta), Recurrence Relations & Master Theorem Rigorous computational complexity theory, asymptotic bounding sets, mathematical induction of recurrence equations, full proof of the Master Theorem and Akra-Bazzi generalization in algorithmic AI design. Academic Learning Outcom...
Priority Queues, Balanced Search Trees (AVL, Red-Black) & Spatial Indexing (KD-Trees)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 10.2: Advanced Data Structures Priority Queues, Balanced Search Trees (AVL, Red-Black) & Spatial Indexing (KD-Trees) Structural invariants, logarithmic height guarantees, tree rotations, Fibonacci heaps, and multidimensional geometric space partitioning with k-d trees for high-dimensional nearest neighbor search. Academic Learning Outcom...
Summative Algorithms Lab: High-Performance Data Structures & Empirical Benchmarking
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 10.3: Practical Lab Studio Summative Algorithms Lab: High-Performance Data Structures & Empirical Benchmarking Applied laboratory engineering probabilistic streaming algorithms, Bloom filters, Count-Min sketches, memory cache-locality profiling, and rigorous statistical benchmarking of algorithmic run-times. Academic Learning Outcomes En...
Object-Oriented Design Patterns (Factory, Strategy, Observer) for AI Pipelines
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 11.1: Software Engineering Object-Oriented Design Patterns (Factory, Strategy, Observer) for AI Pipelines Architectural decoupling, Gang of Four (GoF) structural and behavioral design patterns, abstract base interfaces, dynamic strategy dispatching, and event-driven observer pipelines in production machine learning software. Academic Learnin...
C++ Systems Programming, Memory Management, Pointers & Python C-Extensions (pybind11)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 11.2: Systems Programming C++ Systems Programming, Memory Management, Pointers & Python C-Extensions (pybind11) Low-level systems engineering for AI, C++ memory architecture, RAII, move semantics, zero-copy Python Buffer Protocol, Global Interpreter Lock (GIL) lifecycle, and seamless pybind11 integration. Academic Learning Outcomes Analy...
Summative Software Engineering Lab: Building a C++/Python High-Throughput Matrix Engine
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 11.3: Practical Lab Studio Summative Software Engineering Lab: Building a C++/Python High-Throughput Matrix Engine Applied laboratory engineering a cache-oblivious matrix multiplication engine, loop tiling, memory alignment, OpenMP multi-core parallelization, and empirical performance benchmarking against standard BLAS routines. Academic Lea...
Numerical Stability, Condition Numbers, Floating-Point Precision & Rounding Error Analysis
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 12.1: Numerical Analysis Numerical Stability, Condition Numbers, Floating-Point Precision & Rounding Error Analysis Mathematical error propagation, IEEE 754 floating-point dynamics, Wilkinson's backward error analysis, matrix condition numbers, perturbation bounds, and stabilization of machine learning loss functions. Academic Learning O...
Direct and Iterative Solvers: LU, QR, Cholesky Factorization & Conjugate Gradient Methods
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 12.2: Matrix Factorization Direct and Iterative Solvers: LU, QR, Cholesky Factorization & Conjugate Gradient Methods Algorithmic matrix decompositions, Gaussian elimination with partial pivoting (LUP), Cholesky decomposition of symmetric positive definite systems, Householder QR, and Krylov subspace Conjugate Gradient solvers. Academic L...
Summative Numerical Computing Lab: Solving Large-Scale Sparse Linear Systems for AI
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 12.3: Practical Lab Studio Summative Numerical Computing Lab: Solving Large-Scale Sparse Linear Systems for AI Applied laboratory engineering Compressed Sparse Row (CSR) and Coordinate (COO) matrix storage, high-throughput Sparse Matrix-Vector (SpMV) multiplication, Jacobi preconditioning, and solving massive regularized regression systems....
Relational Schema Design, Normalization, ACID Transactions & SQL for Analytics
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 13.1: Database Systems Relational Schema Design, Normalization, ACID Transactions & SQL for Analytics Relational database theory, functional dependencies, Armstrong's axioms, Boyce-Codd Normal Form (BCNF), ACID concurrency control, Two-Phase Locking, and high-performance analytical SQL for ML feature stores. Academic Learning Outcomes De...
NoSQL Distributed Stores, Columnar Databases & Vector Indexing (HNSW, IVFFlat)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 13.2: Distributed & Vector Stores NoSQL Distributed Stores, Columnar Databases & Vector Indexing (HNSW, IVFFlat) Distributed data architectures, the CAP theorem, columnar OLAP engines, high-dimensional vector search, Product Quantization (PQ), Inverted File Indexing (IVFFlat), and Hierarchical Navigable Small World (HNSW) graphs. Aca...
Summative Data Systems Lab: Building a Vector Database Retrieval Engine with Milvus/Pinecone
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 13.3: Practical Lab Studio Summative Data Systems Lab: Building a Vector Database Retrieval Engine with Milvus/Pinecone Applied laboratory constructing an enterprise-grade vector retrieval service, hybrid sparse-dense scoring, Reciprocal Rank Fusion (RRF), metadata pre-filtering, and distributed index segment lifecycle management. Academic L...
Modern CPU Architecture: Memory Hierarchy, Cache Misses & SIMD Vectorization
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 14.1: Hardware & Parallel Computing Modern CPU Architecture: Memory Hierarchy, Cache Misses & SIMD Vectorization Microarchitectural execution, super-scalar pipelines, out-of-order execution, branch prediction, MESI cache coherence, SIMD vector registers (AVX-512/NEON), and the Roofline Performance Model. Academic Learning Outcomes An...
GPU Architecture, CUDA Programming Model, Thread Hierarchy & Shared Memory Optimization
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 14.2: GPU Architecture & CUDA GPU Architecture, CUDA Programming Model, Thread Hierarchy & Shared Memory Optimization Massively parallel computing, Streaming Multiprocessors (SM), Single Instruction Multiple Threads (SIMT), warp scheduling, memory coalescing, bank conflict minimization, and shared memory tiling. Academic Learning Out...
Summative Parallel Computing Lab: Writing and Profiling Custom CUDA Kernels for Tensor Ops
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 14.3: Practical Lab Studio Summative Parallel Computing Lab: Writing and Profiling Custom CUDA Kernels for Tensor Ops Applied laboratory engineering high-performance CUDA kernels, warp shuffle reduction primitives (__shfl_down_sync), fused LayerNorm operators, and profiling with NVIDIA Nsight Compute. Academic Learning Outcomes Engineer warp...
Projective Geometry, Camera Calibration, Epipolar Geometry & Stereoscopic Vision
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 15.1: Visual Geometry Projective Geometry, Camera Calibration, Epipolar Geometry & Stereoscopic Vision Projective spaces \mathbb{P}2 and \mathbb{P}3, pinhole camera modeling, intrinsic/extrinsic calibrations, Zhang's method, the Essential and Fundamental matrices, epipolar lines, and stereoscopic depth triangulation. Academic Learning Ou...
Deep Convolutional Backbones: ResNet, EfficientNet, ConvNeXt & Feature Pyramid Networks
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 15.2: Deep Vision Backbones Deep Convolutional Backbones: ResNet, EfficientNet, ConvNeXt & Feature Pyramid Networks Evolution of convolutional representation learning, degradation dynamics, residual skip connections, compound coefficient scaling, modern depthwise ConvNeXt architectures, and multi-scale Feature Pyramid Networks (FPN). Aca...
Summative Vision Lab: Implementing Feature Extraction, Homography & 3D Point Cloud Processing
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 15.3: Practical Lab Studio Summative Vision Lab: Implementing Feature Extraction, Homography & 3D Point Cloud Processing Applied laboratory engineering scale-space keypoint detectors, RANSAC planar homography estimation, image stitching, and 3D point cloud backprojection from stereo depth maps. Academic Learning Outcomes Implement Differ...
Anchor-Based and Anchor-Free Detection: Faster R-CNN, SSD, RetinaNet & YOLOv8/v10
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 16.1: Object Detection Anchor-Based and Anchor-Free Detection: Faster R-CNN, SSD, RetinaNet & YOLOv8/v10 Two-stage Region Proposal Networks (RPN), single-stage dense object detectors, addressing extreme foreground-background class imbalance via Focal Loss, and NMS-free anchor-free YOLO architectures. Academic Learning Outcomes Formulate...
Semantic and Instance Segmentation: U-Net, Mask R-CNN & Segment Anything (SAM)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 16.2: Image Segmentation Semantic and Instance Segmentation: U-Net, Mask R-CNN & Segment Anything (SAM) Pixel-level dense prediction, encoder-decoder architectures, skip-connection feature reassembly, instance mask prediction heads, and the Segment Anything Model (SAM) foundation architecture. Academic Learning Outcomes Distinguish seman...
Summative Detection Lab: Custom Multi-Scale Object Detector Training & mAP Profiling
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 16.3: Practical Lab Studio Summative Detection Lab: Custom Multi-Scale Object Detector Training & mAP Profiling Applied laboratory engineering multi-scale detection loss functions (Complete IoU / CIoU), mosaic data augmentation, and implementing full COCO Mean Average Precision (mAP@[.5:.05:.95]) evaluation suites. Academic Learning Outc...
Formal Grammars, Dependency Parsing, Morphological Analysis & Language Models (N-gram)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 17.1: NLP Foundations Formal Grammars, Dependency Parsing, Morphological Analysis & Language Models (N-gram) Chomsky hierarchy of formal grammars, context-free parsing via the Cocke-Younger-Kasami (CKY) dynamic programming algorithm, transition-based dependency parsing, and statistical N-gram language modeling with Kneser-Ney smoothing....
Dense Semantic Embeddings: Word2Vec, GloVe, FastText & Subword Tokenization (BPE, WordPiece)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 17.2: Vector Semantics Dense Semantic Embeddings: Word2Vec, GloVe, FastText & Subword Tokenization (BPE, WordPiece) Continuous vector space semantics, the Distributional Hypothesis, Word2Vec Skip-Gram with Negative Sampling (SGNS), global co-occurrence factorization in GloVe, subword n-gram hashing in FastText, and subword tokenization (...
Summative NLP Lab: Building an End-to-End Neural POS Tagger and Named Entity Recognizer
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 17.3: Practical Lab Studio Summative NLP Lab: Building an End-to-End Neural POS Tagger and Named Entity Recognizer Applied laboratory constructing deep sequence labeling systems, Bidirectional LSTM encoders, Linear-Chain Conditional Random Field (CRF) transitions, and Viterbi dynamic programming decoding. Academic Learning Outcomes Formulate...
Description Logics, Web Ontology Language (OWL) & Resource Description Framework (RDF)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 18.1: Semantic Web & Ontologies Description Logics, Web Ontology Language (OWL) & Resource Description Framework (RDF) Knowledge representation foundations, RDF graph models, Description Logics (\mathcal{ALC}, \mathcal{SROIQ}), OWL 2 DL profiles, model-theoretic semantics, TBox/ABox structures, and Tableaux reasoning algorithms. Acad...
Knowledge Graph Construction, Relation Extraction & Knowledge Graph Embeddings (TransE, RotatE)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 18.2: Knowledge Graph Embeddings Knowledge Graph Construction, Relation Extraction & Knowledge Graph Embeddings (TransE, RotatE) Knowledge graph engineering, distant supervision relation extraction, representation learning on multirelational graphs, translational models (TransE, TransH), and complex rotational manifolds (RotatE). Academi...
Summative Knowledge Graph Lab: Building and Querying an Enterprise Knowledge Graph with SPARQL
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 18.3: Practical Lab Studio Summative Knowledge Graph Lab: Building and Querying an Enterprise Knowledge Graph with SPARQL Applied laboratory engineering persistent RDF triplestores, Hexastore and permutation index architectures (SPO, POS, OSP), and writing optimized SPARQL 1.1 graph queries for neuro-symbolic reasoning. Academic Learning Out...
Latent Variable Models, Variational Autoencoders (VAEs) & Reparameterization Dynamics
1. Introduction: The Deep Generative Modeling Paradigm and Latent Variable Formulations Modern machine learning distinguishes fundamentally between discriminative modeling—which seeks to capture the conditional distribution \\(p(y|x)\\) mapping high-dimensional observations \\(x \in \mathcal{X}\\) to targets \\(y \in \mathcal{Y}\\)—and generative modeling, which models the underlying data-generati...
Generative Adversarial Networks (GANs): Minimax Game, Wasserstein GAN & Spectral Normalization
1. Introduction: Game-Theoretic Foundations of Adversarial Learning Unlike Variational Autoencoders that optimize an explicit surrogate density lower bound (ELBO), Generative Adversarial Networks (GANs), introduced by Goodfellow et al. (2014), abandon explicit likelihood formulation entirely. GANs cast generative modeling as an adversarial, two-player zero-sum game between two competing neural net...
Summative Generative Lab: Training a High-Fidelity WGAN-GP on Complex Image Distributions
1. Laboratory Overview: Rigorous Engineering of Adversarial Deep Generators This summative laboratory provides an industrial-grade, end-to-end framework for training, stabilizing, and quantitatively evaluating Generative Adversarial Networks. Generative modeling projects in production environments frequently fail due to unmonitored gradient explosions, subtle implementation bugs in gradient penalt...
Forward and Reverse Markov Chains in Diffusion: DDPM Math, Noise Schedules & Loss Objectives
1. Introduction: Generative Modeling as Inverting Physical Diffusion Processes While Variational Autoencoders compress images into bottleneck latents and Generative Adversarial Networks engage in unstable minimax games, Denoising Diffusion Probabilistic Models (DDPM), pioneered by Sohl-Dickstein et al. (2015) and modernized by Ho, Jain & Abbeel (2020), formulate generation as the progressive inver...
Score-Based Generative Models, Continuous-Time SDEs & Fast Sampling (DDIM, DPM-Solver)
1. Introduction: Unifying Score Matching and Diffusion via Continuous-Time SDEs While discrete DDPM models demonstrate extraordinary image synthesis quality, their theoretical formulation remained somewhat disparate from another major class of generative models: Score-Based Generative Models (Song & Ermon, 2019). Song et al. (2020) resolved this divide by proving that both DDPMs and score-based mo...
Summative Diffusion Lab: Training a Latent Diffusion Model with Classifier-Free Guidance
1. Laboratory Overview: Engineering High-Fidelity Latent Diffusion Systems Direct pixel-space diffusion models, while theoretically powerful, suffer from catastrophic computational demands: evaluating a high-resolution U-Net across 512x512 RGB tensors over hundreds of timesteps requires immense GPU memory and massive FLOP budgets. Rombach et al. (2022) solved this bottleneck with Latent Diffusion...
Kinematics and Dynamics of Mobile Robots, Differential Drive & Ackermann Steering
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 21.1: Mobile Robot Kinematics Kinematics and Dynamics of Mobile Robots, Differential Drive & Ackermann Steering Rigid body transformations in SE(2), non-holonomic motion constraints, forward and inverse differential drive kinematics, Ackermann steering geometry, wheel slip dynamics, and odometric covariance propagation. Academic Learning...
Simultaneous Localization and Mapping (SLAM): Extended Kalman Filter (EKF-SLAM) & Graph-SLAM
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 21.2: Spatial AI & SLAM Simultaneous Localization and Mapping (SLAM): Extended Kalman Filter (EKF-SLAM) & Graph-SLAM Probabilistic state estimation, landmark-based EKF-SLAM, measurement models, loop closure detection, factor graph optimization, Gauss-Newton solvers, and sparse Cholesky factorization. Academic Learning Outcomes Formul...
Summative Robotics Lab: ROS 2 Navigation Stack Implementation with Costmaps & Path Planning (A*)
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 21.3: Autonomous Systems Lab Summative Robotics Lab: ROS 2 Navigation Stack Implementation with Costmaps & Path Planning (A*) ROS 2 DDS architecture, managed lifecycle nodes, Nav2 Costmap 2D multi-layer inflation dynamics, A* global search with Octile heuristics, and gradient descent path smoothing. Academic Learning Outcomes Engineer ma...
Model Lineage, Versioning (DVC), Automated Testing & CI/CD Pipelines for Machine Learning
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 22.1: MLOps Foundations Model Lineage, Versioning (DVC), Automated Testing & CI/CD Pipelines for Machine Learning Data Version Control (DVC), content-addressable storage, behavioral testing of deep models (invariance and directional tests), and continuous machine learning (CML) integration. Academic Learning Outcomes Analyze the software...
High-Performance Model Serving: Triton Inference Server, vLLM & Low-Latency API Gateways
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 22.2: Production Inference Systems High-Performance Model Serving: Triton Inference Server, vLLM & Low-Latency API Gateways Hardware Roofline analysis, dynamic batching algorithms, multi-model GPU concurrency, PagedAttention KV-cache management in vLLM, and asynchronous gRPC microservice gateways. Academic Learning Outcomes Analyze deep...
Summative MLOps Studio: Deploying an Auto-Scaling Inference Microservice with Canary Rollouts
B.Sc. Artificial Intelligence NUC CCMAS Benchmark Module 22.3: Production MLOps Studio Summative MLOps Studio: Deploying an Auto-Scaling Inference Microservice with Canary Rollouts Kubernetes GPU deployment, Horizontal Pod Autoscaler (HPA), Prometheus custom telemetry metrics, Istio service mesh traffic splitting, and automated Population Stability Index (PSI) drift rollbacks. Academic Learning Ou...
Formulating Novel AI Research Questions, Literature Review Synthesis & Hypothesis Testing
1. Introduction: Epistemological Foundations and the Nature of AI Scientific Inquiry Artificial Intelligence occupies a unique intersection within the philosophy of science: it is simultaneously an engineering discipline (synthesizing computational systems to perform intelligent tasks) and an empirical empirical science (observing, measuring, and explaining emergent computational phenomena within...
Experimental Design, Statistical Significance (Bootstrap, Wilcoxon) & Reproducibility Standards
1. Introduction: The Reproducibility Crisis in Machine Learning and AI Over the past decade, the rapid pace of artificial intelligence publications has encountered an acute crisis of empirical reproducibility. Henderson et al. (2018) demonstrated in reinforcement learning that running the identical algorithm with identical hyperparameters on identical benchmarks yields non-overlapping performance...
Summative Research Seminar: Crafting an Archival-Quality IEEE/NeurIPS Conference Paper
1. Seminar Overview: The Anatomy of Tier-1 Academic Publications in AI Publishing in flagship artificial intelligence venues (NeurIPS, ICML, ICLR, CVPR, ACL, IEEE TPAMI) requires more than novel code or high benchmark accuracy: it demands clear mathematical formulation, rigorous empirical contextualization, transparent threat modeling, and professional academic prose. Reviewers at top-tier confere...
Capstone Project Problem Formulation, Architectural Blueprint & Feasibility Study
1. Introduction: The Capstone Undergraduate AI Dissertation Mandate The undergraduate AI Senior Capstone Project represents the culmination of academic study, synthesizing theoretical foundations (linear algebra, probability, optimization, formal logic) with practical software engineering and machine learning systems design. Unlike routine homework assignments or standalone classroom labs, the Cap...
Comprehensive Empirical Validation: Benchmark Performance, Ablation Studies & Failure Modes
1. Introduction: The Experimental Mandate of the Capstone Dissertation The empirical evaluation chapter constitutes the scientific core of the undergraduate AI dissertation. A capstone project cannot rely on unsubstantiated assertions of superiority; every claim must be established through methodical empirical experimentation against credible baselines under controlled conditions. A comprehensive...
Summative Capstone Defense: Formal Academic Monograph, Code Artifacts & Oral Defense
1. Introduction: The Summative Capstone Examination and Academic Defense The final milestone of the Bachelor of Science in Artificial Intelligence degree is the formal Capstone Examination. The candidate must present and defend their complete body of work before an academic jury comprising the primary research supervisor, departmental faculty assessors, and an external examiner appointed by the Un...
Academic Ecosystem & Connected Pathways
Deepen your studies across connected learning pathways in the PAMCET academic network:
Instructor
PAMCET Learning Team
Institutional Course Curation · PAMCET
Qualifications:
PAMCET Learning Team — Institutional Course Curation, Digitpen Hub LtdMy Skills:
- Curriculum Curation
- Course Content Review
- Learning Outcome Design
- Course Publishing & Quality Assurance
- Platform Content Operations
Reviews
Course Includes
Download Syllabus & Study Guide
Receive the full lecture syllabus, reading list, and milestone schedule for Bachelor of Science (B.Sc.) in Artificial Intelligence: Algorithmic Foundations, Autonomous Agents & Machine Intelligence.
By submitting, you agree to receive educational updates. You can unsubscribe anytime. Read our Privacy Policy.

