TY AI & DS Specialization & Advanced Architectures

Statistical Machine Learning, Deep Neural Networks, and Distributed Cloud Computing

The Third Year (TY) track represents the specialization apex of the undergraduate AI & DS program. Students apply core mathematical and data structure models to advanced statistical learning techniques, deep neural net optimization, and large-scale data stream engines.

Advanced Specialization Modules

The TY academic curriculum encompasses four critical pillars of modern data engineering:

Third Year Core Specialization Pillars
  • Statistical Machine Learning (SML): Supervised learning models (SVMs, Decision Forests, Gradient Boosting), unsupervised clustering (K-Means, DBSCAN), regularization techniques ($L_1$ Lasso, $L_2$ Ridge), and bias-variance tradeoff evaluation.
  • Deep Neural Networks & Architectures: Backpropagation calculus, Convolutional Neural Networks (CNNs) for spatial vision, Recurrent Neural Networks (RNNs/LSTMs) for sequence modeling, and Transformer attention mechanisms.
  • Big Data Distributed Systems: MapReduce paradigm, Apache Spark resilient distributed datasets (RDDs), distributed storage (HDFS), and stream processing topologies.
  • Cloud Platform Infrastructure: Containerization (Docker, Kubernetes), serverless computing, microservice deployment, and IAM policy security frameworks.

Repository Indexing & Peer Review Pipeline

Study materials, practical lab codes, and comprehensive question banks for Third Year modules are currently under active curation by Sarthak Pawar in coordination with academic faculty. Verified resources will be uploaded as structured direct downloads as semester sessions commence.

Academic Note: For foundational linear algebra and calculus concepts essential to understanding Machine Learning loss functions, please refer to our FY Mathematics Repository.