Machine Learning
Flex Analytics: AI & Machine Learning Professional Program Syllabus.
Module 1: Foundations of AI & Python Programming
Introduction to Artificial Intelligence (AI)
What is AI? History and Evolution.
Types of AI: ANI (Narrow), AGI (General), ASI (Super).
Branches of AI: Machine Learning, Deep Learning, NLP, Computer Vision, Robotics, Expert Systems.
Real-world Applications and Impact of AI.
Ethical Considerations in AI.
Introduction to Machine Learning (ML)
What is Machine Learning? Relationship to AI and Data Science.
Types of Machine Learning:
Supervised Learning (Regression, Classification)
Unsupervised Learning (Clustering, Dimensionality Reduction)
Reinforcement Learning
The ML Workflow: Problem Definition, Data Collection, Data Preprocessing, Model Building, Evaluation, Deployment.
Essential Mathematics for ML
Linear Algebra: Vectors, Matrices, Operations, Dot Product, Eigenvectors & Eigenvalues (Conceptual).
Calculus: Derivatives, Gradients, Partial Derivatives, Chain Rule (Conceptual understanding for Gradient Descent).
Probability & Statistics: Basic Probability, Conditional Probability, Bayes’ Theorem, Random Variables, Probability Distributions (Normal, Binomial), Descriptive Statistics (Mean, Median, Mode, Variance, Std Dev), Hypothesis Testing (Conceptual).
Python for Data Science & ML
Python Basics: Syntax, Data Types, Variables, Operators, Control Flow (if/else, loops).
Data Structures: Lists, Tuples, Dictionaries, Sets.
Functions, Modules, and Packages.
Object-Oriented Programming (OOP) Concepts in Python (Classes, Objects).
NumPy: Array creation, Indexing, Slicing, Mathematical operations, Linear algebra functions.
Pandas: Series, DataFrames, Data loading (CSV, Excel, SQL), Data inspection, Cleaning (Missing values, Duplicates), Manipulation (Filtering, Sorting, Grouping, Merging), Data Transformation.
Matplotlib & Seaborn: Data Visualization (Line plots, Scatter plots, Bar charts, Histograms, Box plots, Heatmaps).
Module 2: Core Machine Learning Algorithms .
Data Preprocessing & Feature Engineering
Handling Missing Data (Imputation techniques).
Encoding Categorical Data (One-Hot Encoding, Label Encoding).
Feature Scaling (Normalization, Standardization).
Feature Selection Techniques (Filter, Wrapper, Embedded methods – conceptual).
Feature Engineering: Creating new features from existing ones.
Train-Test Split & Cross-Validation.
Supervised Learning – Regression
Simple Linear Regression.
Multiple Linear Regression.
Polynomial Regression.
Assumptions of Linear Regression.
Regularization Techniques: Ridge (L2) & Lasso (L1) Regression.
Evaluation Metrics: MAE, MSE, RMSE, R-squared, Adjusted R-squared.
Hands-on: Implementing regression models on datasets.
Supervised Learning – Classification
Logistic Regression.
K-Nearest Neighbors (KNN).
Support Vector Machines (SVM) & Kernels.
Decision Trees (CART, ID3, C4.5 concepts).
Naive Bayes Classifiers (Gaussian, Multinomial).
Evaluation Metrics: Accuracy, Precision, Recall, F1-Score, Confusion Matrix, ROC Curve & AUC.
Handling Imbalanced Classes (SMOTE – conceptual).
Hands-on: Implementing classification models, interpreting results.
Ensemble Learning
Concept of Ensemble Learning: Bagging & Boosting.
Random Forest (Bagging).
Boosting Algorithms: AdaBoost, Gradient Boosting, XGBoost, LightGBM, CatBoost (overview and practical use of at least one).
Stacking (Conceptual).
Hands-on: Implementing and comparing ensemble models.
Unsupervised Learning
Clustering:
K-Means Clustering (Elbow method, Silhouette score).
Hierarchical Clustering (Agglomerative, Divisive, Dendrograms).
DBSCAN.
Dimensionality Reduction:
Principal Component Analysis (PCA).
t-distributed Stochastic Neighbor Embedding (t-SNE) for visualization (conceptual).
Association Rule Mining (Apriori Algorithm – conceptual).
Hands-on: Implementing clustering and PCA.
Model Evaluation, Validation & Improvement
Bias-Variance Trade-off.
Overfitting and Underfitting: Detection and Mitigation.
Cross-Validation Techniques (K-Fold, Stratified K-Fold).
Hyperparameter Tuning: Grid Search, Random Search, Bayesian Optimization (conceptual).
Module 3: Deep Learning .
Introduction to Neural Networks & Deep Learning
Biological Neuron vs. Artificial Neuron (Perceptron).
Activation Functions (Sigmoid, Tanh, ReLU, Leaky ReLU, Softmax).
Multi-Layer Perceptrons (MLPs) / Feedforward Neural Networks.
Forward Propagation & Backpropagation Algorithm.
Loss Functions (MSE, Cross-Entropy).
Optimizers (SGD, Momentum, AdaGrad, RMSProp, Adam).
Initialization Techniques.
Regularization in NNs (Dropout, L1/L2, Early Stopping).
Batch Normalization.
Deep Learning Frameworks
Introduction to TensorFlow & Keras (or PyTorch – choose one as primary).
Building Sequential and Functional Models.
Compiling, Training, Evaluating, and Predicting with NNs.
Saving and Loading Models.
Callbacks.
Hands-on: Building and training basic NNs for classification and regression.
Convolutional Neural Networks (CNNs)
The Convolution Operation.
Pooling Layers (Max, Average).
CNN Architectures (LeNet, AlexNet, VGG, ResNet, Inception – overview).
Applications: Image Classification, Object Detection basics.
Transfer Learning & Fine-tuning pre-trained models.
Data Augmentation for Images.
Hands-on: Building CNNs for image classification (e.g., CIFAR-10, MNIST).
Recurrent Neural Networks (RNNs)
Concept of Sequential Data.
Basic RNN structure and limitations (Vanishing/Exploding Gradients).
Long Short-Term Memory (LSTM) Networks.
Gated Recurrent Units (GRUs).
Applications: Time Series Analysis, Text Generation, Sentiment Analysis.
Hands-on: Building RNNs/LSTMs for sequence modeling tasks.
Module 4: Specialized AI Applications.
Natural Language Processing (NLP)
Text Preprocessing: Tokenization, Stemming, Lemmatization, Stop Word Removal.
Text Representation:
Bag-of-Words (BoW).
TF-IDF (Term Frequency-Inverse Document Frequency).
Word Embeddings (Word2Vec, GloVe, FastText – conceptual and usage).
Applications:
Sentiment Analysis.
Text Classification.
Topic Modeling (LDA – conceptual).
Introduction to Sequence-to-Sequence Models & Attention Mechanism.
Transformers (BERT, GPT – high-level overview and usage via libraries like Hugging Face).
Hands-on: NLP tasks like sentiment analysis or text classification.
Computer Vision (CV) – Deeper Dive
Image Processing Basics (Filters, Edge Detection, Transformations).
Object Detection Algorithms (YOLO, SSD – overview and usage of pre-trained models).
Image Segmentation (Conceptual).
Generative Adversarial Networks (GANs) for Image Generation (Conceptual and simple examples).
Hands-on: Using pre-trained models for object detection or exploring GANs.
Reinforcement Learning (RL) – Introduction
Core Concepts: Agent, Environment, State, Action, Reward, Policy.
Markov Decision Processes (MDPs).
Value-based Methods (Q-Learning, Deep Q-Networks – DQN).
Policy-based Methods (Policy Gradients – conceptual).
Applications (Games, Robotics, Recommendation Systems).
Hands-on (Optional/Conceptual): Simple RL problem using libraries like OpenAI Gym.
Module 5: Model Deployment & MLOps
Machine Learning Operations (MLOps) Principles
What is MLOps? Importance and Lifecycle.
Version Control (Git/GitHub for code and models – DVC/MLflow concepts).
Experiment Tracking (MLflow or similar).
Model Deployment Strategies
Saving and Loading Models (Pickle, Joblib, framework-specific).
Creating APIs for ML Models (Flask/FastAPI).
Containerization with Docker (Basics).
Deployment to Cloud Platforms (Overview of AWS SageMaker, Google AI Platform, Azure ML).
Serverless Deployment (AWS Lambda, Google Cloud Functions – conceptual).
Hands-on: Deploying a simple model as a REST API using Flask/FastAPI.
Model Monitoring & Maintenance
Monitoring Model Performance in Production.
Concept Drift and Data Drift.
Retraining Strategies.
Feedback Loops.
Ethical AI & Responsible AI Development
Bias in AI: Sources, Detection, Mitigation techniques.
Fairness, Accountability, Transparency (FAT/FATE).
Explainable AI (XAI): LIME, SHAP (conceptual and basic usage).
Data Privacy and Security in ML.
Module 6: Career Preparation
Capstone Project
Students work individually or in small groups on a real-world or simulated problem.
End-to-end execution: Problem definition, data collection/sourcing, preprocessing, model selection & training, evaluation, and (optionally) basic deployment/presentation.
Guidance and mentorship from instructors.
Project Presentation and Report.
Career Development & Portfolio Building
Building a strong Data Science/ML portfolio (GitHub).
Resume and LinkedIn Profile Optimization.
Interview Preparation: Technical and Behavioral questions.
Staying Updated: Resources, Communities, Continuous Learning.
Tools & Technologies Covered:
Programming Language: Python
Core Libraries: NumPy, Pandas, Matplotlib, Seaborn, Scikit-learn
Deep Learning Frameworks: TensorFlow & Keras (or PyTorch)
NLP Libraries (Optional): NLTK, SpaCy, Hugging Face Transformers
Deployment Tools (Conceptual/Basic): Flask/FastAPI, Docker
Version Control: Git, GitHub
IDE/Environment: Jupyter Notebooks, Google Colab, VS Code
Prerequisites:
Basic understanding of programming concepts (any language is a plus, but Python will be taught).
Logical and analytical thinking skills.
High school level mathematics.
