PAHSU · Final Year B.Tech. Computer Science and Engineering · NEP 2020
Artificial Intelligence and Machine Learning
CSEPCC-13 · Semester VII · 3 credits · ISE 30 + ESE 70
Unit 1: Introduction to AI and ML
7 hoursDefinitions, history and evolution of AI; how AI, machine learning and deep learning relate and differ; supervised, unsupervised and reinforcement learning; real-world applications in healthcare, finance, robotics and NLP; ethics in AI: bias, fairness and explainability.
Examined in the 70-mark end-semester paper
- Foundations and history of AI
- AI, ML, deep learning and data science
- Introduction to machine learning
- Types of learning
- Explainable and responsible AI
Unit 2: Data Pre-processing and Processing
6 hoursTypes of data; handling missing values, duplicates and outliers; min-max normalisation and z-score standardisation; one-hot and label encoding; training, validation and test splits; exploratory data analysis with visualisation and correlation.
Examined in the 70-mark end-semester paper
- Data wrangling and preprocessing
- Feature engineering
- Cross-validation and micro and macro averaging
- Univariate and multivariate visualisation
- Correlation
Unit 3: Machine Learning Algorithms
7 hoursSimple and multiple linear regression, the cost function and gradient descent; decision trees: entropy, information gain, Gini impurity and pruning; random forests, ensembles, bagging and bootstrapping; support vector machines: hyperplanes, margins and the kernel trick; k-nearest neighbours: distance metrics and choosing k.
Examined in the 70-mark end-semester paper
- Linear and polynomial regression
- Gradient descent
- Decision trees and CART
- Bagging and random forests
- Support vector machines and kernels
- K-nearest neighbours
Unit 4: Introduction to Neural Networks and Model Evaluation
8 hoursThe perceptron; single- and multi-layer perceptrons and their layers; sigmoid and ReLU activations, forward propagation and the basics of backpropagation; classification, regression and prediction problems; the confusion matrix, accuracy, precision, recall and F1-score.
Examined in the 70-mark end-semester paper
- Perceptron and multilayer perceptron
- Activation functions
- Backpropagation learning
- Confusion matrix, precision, recall and F1
Unit 5: Introduction to Deep Learning
7 hoursThe need for deep learning on high-dimensional data; vanishing gradients and overfitting in deep networks; stochastic gradient descent and Adam; dropout and L1/L2 regularisation; an introduction to CNNs.
Examined in the 70-mark end-semester paper
- Deep learning vs machine learning
- Gradient descent and vanishing and exploding gradients
- Optimisers: momentum, Nesterov, AdaGrad, RMSProp and Adam
- Regularisation: dropout, batch normalisation and early stopping
- Introduction to CNNs and RNNs
Unit 6: Applications of Deep Learning
7 hoursApplication-oriented deep architectures: U-Net for medical image segmentation, ResNet for deep image classification and GoogLeNet for efficient vision models; segmentation, clustering and regression prediction; testing and validation approaches.
Examined in the 70-mark end-semester paper
- CNN architectures: VGG, GoogLeNet, ResNet and DenseNet
- Object detection and image segmentation
- Regression evaluation metrics
Unit contents are summarised from the official PAHSU syllabus . AiCanCode.org is not affiliated with Punyashlok Ahilyadevi Holkar Solapur University (PAHSU).