Neeraj Baghel

Course Information: AI&ML (CSET301)

Detailed Syllabus

Module 1: Designing a learning system life cycle, Types of machine learning: Supervised learning, Unsupervised learning, Reinforcement learning, Introduction to Linear Regression and Classification model, Applications Gradient Descent Algorithm: Learning Algorithm, Linear regression using gradient descent, Learning rate, Polynomial Regression, Regularization (Ridge, Lasso, Elastic), Regression (MSE, MAE, RMSE, R2 Score), Performance Metrics: Classification (Confusion Matrix, Accuracy, Precision, Recall, F1-score, ROC-AUC).

Module 2: Ensemble Learning, Bagging, Bootstrap and Aggregation, Random Forest. Boosting, AdaBoost, Gradient Boost. Hyperparameter Tunning, Methods for Hyperparameter Tuning: Grid Search and Random Search. Unsupervised learning (clustering, Association rule learning, Dimensionality reduction), Common Distance Measures, Hierarchical Clustering – agglomerative and divisive, Dendrogram, Similarity measures for hierarchical clustering, DBSCAN, Cluster Quality (R index, Silhouette Coefficient), Dimensionality Reduction, Principal Component Analysis, Overfitting and Underfitting, Bias and Variance, Cross Validation.

Module 3: Artificial Neural Network, Neural network representation, Perceptron model, Stepwise v/s Sigmoid function, Multilayer perceptron model, Backpropagation Algorithm, Activation Functions, Stochastic Gradient Descent, Batch Gradient Descent, Mini-Batch Gradient Descent, Vanishing and Exploding Gradients, Overfitting Problem, Dropout and Early Stopping. Computer Vision, Convolutional Neural Networks,

Module 4: Deep Learning for Sequential Data, Recurrent Neural Networks, LSTM, GRU, Natural Language Processing, Stemming, Lemmatization, Tokenization, feature engineering, Bag of Words, TFIDF, Word Embeddings, Transformers (BERT and GPT)

Module 5: Population-based algorithms: Genetic Algorithm, Fitness Function, Selection, Crossover, Mutation, Swarm Optimization, Particle Swarm Optimization, Ant-Colony Optimization, Reinforcement Learning, Actors, State, Reward Policy, and Actions.

Module 1:

Introduction to AIML [Link]
LifeCycle: [Link]
Types of ML: [Link]
Regression: [Link]
Performance & Regularization: [Link]

Module 2:

Ensemble Learning [Link]
Clustering [Link]
Dimensionality Reduction [Link]

Module 3:

ANN [Link]
CNN [Link]
Gradient Descent Types [Link]

Module 4:

RNN [Link]
NLP [Link]
Transformer [Link]

Course Certification:

Deep Learning Specialization [Link]
Neural Networks and Deep Learning:[Link]
Convolutional Neural Networks: [Link]
Improving Deep Neural Networks: [Link]
Introduction to TensorFlow [Link]

Evaluation policy:

Mid Semester Examination (20 Marks)
End Semester Examination (40 Marks)
Lab Continuous Evaluation (20 Marks)
Certification (5 Marks)
Project (15 Marks)

Lab Experiments List:

Linear and Polynomial Regression
Ridge, Lasso, Elastic Regression
AdaBoost Classification and Cross Validation
Agglomerative, DBSCAN Clustering
Principal Component Analysis
Artificial Neural Network
Hyperparameter Tuning
Convolutional Neural Networks
Recurrent Neural Networks
LSTM and GRU
Trasformers, BERT
Genetic Algorithm

COURSE SUMMARY

The objective of this course is to provide a rigorous training on the fundamental concepts, algorithms, and theories in artificial intelligence and machine learning. The lectures will cover various supervised and unsupervised learning algorithms that will help students to design and develop AI systems to solve real-world problems.

COURSE-SPECIFIC LEARNING OUTCOMES (CO)

CLO1: To articulate various supervised and unsupervised machine learning approaches.

CLO2: To build and evaluate models generated from data

CLO3: To implement systems to solve real life problem using AI and Machine Learning approaches.

TEXTBOOKS/LEARNING RESOURCES:

1) T. M. Mitchell, Machine Learning, McGraw Hill, 2017. ISBN-13: 978-1259096952.

2) E. Alpaydin, Introduction to Machine Learning (4th Edition), Phi, 2020. ISBN:13- 978-8120350786.

REFERENCE BOOKS/LEARNING RESOURCES:

1) Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT press, 2016.

Course Instructors:

Tarandeep Singh, Manish Raj, Nitin Arvind Shelke, Akhilesh Rawat, Neeraj Baghel, Yajnaseni Dash, Apeksha Koul, Hoor Fatima, Shwetang Dubey, Pooja Chaudary

Page updated

Google Sites

Report abuse