Semester-wise syllabus for an M.Tech in Machine Learning

 

Semester 1:

 Foundational Concepts 

1. Mathematics for Machine Learning 

   - Linear algebra, calculus, probability, statistics, and optimization (gradient descent, convexity). 

2. Programming for Data Science 

   - Python/R programming, data structures, libraries (NumPy, Pandas, Scikit-learn), and version control (Git). 

3. Introduction to Machine Learning 

   - Supervised/unsupervised learning (regression, classification, clustering), evaluation metrics, bias-variance tradeoff. 

4. Data Acquisition and Preprocessing 

   - Data cleaning, feature engineering, dimensionality reduction (PCA, t-SNE), and handling imbalanced data. 

5. Lab Work 

   - Hands-on projects: EDA, basic ML models (k-NN, decision trees), and Kaggle-style competitions. 

 

Semester 2:

 Core Machine Learning & Advanced Topics 

1. Advanced Machine Learning 

   - Ensemble methods (Random Forest, XGBoost), SVM, Bayesian networks, and probabilistic graphical models. 

2. Deep Learning Fundamentals 

   - Neural networks (CNNs, RNNs), backpropagation, regularization, PyTorch/TensorFlow frameworks. 

3. Big Data Technologies 

   - Distributed computing (Hadoop, Spark), NoSQL databases, and cloud platforms (AWS, GCP). 

4. Optimization for ML 

   - Stochastic optimization, hyperparameter tuning, AutoML, and metaheuristics. 

5. Elective 1 

   - Options: Natural Language Processing (NLP), Computer Vision, Reinforcement Learning. 

6. Lab Work 

   - Implementing CNNs/RNNs, Spark-based data pipelines, and hyperparameter optimization (Optuna, Keras Tuner). 

 

Semester 3:

Specialization & Research 

1. Advanced Deep Learning 

   - Transformers, GANs, attention mechanisms, self-supervised learning, and transfer learning. 

2. Elective 2 

   - Options: Time Series Analysis, Graph Neural Networks, Edge AI. 

3. Elective 3 

   - Options: Explainable AI (XAI), Generative Models, AI Ethics. 

4. Research Project (Phase 1) 

   - Problem formulation, literature review, and experimental setup (e.g., building a recommendation system or medical diagnosis model). 

5. Lab Work 

   - Transformer-based NLP tasks (BERT, GPT), GANs for image generation, edge deployment (TensorFlow Lite). 

 

Semester 4:

Thesis & Industry Applications 

1. Dissertation/Thesis 

   - Focus areas: AI ethics, domain-specific applications (healthcare, finance), or novel algorithms. 

2. Industry Internship (Optional) 

   - Collaborations with tech firms, startups, or research labs (e.g., deploying ML models in production). 

3. Emerging Topics Seminar 

   - Topics: Federated Learning, Quantum Machine Learning, AI for Sustainability. 

4. Seminar & Viva Voce 

   - Presentation and defense of thesis work, peer reviews, and industry feedback. 

 

Electives (Across Semesters 2–3) 

- Natural Language Processing (NLP): Word embeddings, sequence-to-sequence models, sentiment analysis. 

- Computer Vision: Object detection (YOLO), segmentation (U-Net), video analytics. 

- Reinforcement Learning: Q-learning, policy gradients, multi-agent systems. 

- AI in Healthcare: Medical imaging, predictive diagnostics, wearable data analysis. 

- MLOps: Model deployment (Docker, Kubernetes), monitoring, CI/CD pipelines. 

 

Tools & Technologies 

- Frameworks: PyTorch, TensorFlow, Keras, Hugging Face, OpenCV. 

- Cloud Platforms: AWS SageMaker, Google AI Platform, Azure ML. 

- Big Data Tools: Apache Spark, Dask, Kafka. 

- Visualization: Tableau, Matplotlib, Seaborn, Plotly. 

- Deployment: Flask/Django APIs, ONNX, TensorFlow Serving. 

 

Industry Applications 

- Tech: Recommendation systems (Netflix, Amazon), fraud detection. 

- Healthcare: Predictive diagnostics, drug discovery. 

- Finance: Algorithmic trading, credit scoring. 

- Autonomous Systems: Self-driving cars, robotics. 

- Sustainability: Climate modeling, energy optimization.