Deep Learning for AI: Mastering Neural Networks

Deep Learning for AI: Mastering Neural Networks
(In Association with iHUB Divyasampark IIT Roorkee)

Why this Course:

Deep Learning (DL) has revolutionized industries due to its ability to effectively predict outcomes from large Datasets and make complex predictions or decisions. Ranging from the domains like Computer Vision, Natural Language processing, Speech Recognition to the sectors of Healthcare, Finance, e-commerce; DL techniques powers innovation in different sectors. Therefore, this course is meticulously designed to take participants from the fundamentals of Deep Learning to advanced concepts. Starting from Python basics and important Python libraries used for DL, there will be a deep dive into the use of deep neural networks to solve complex problems and make predictions. So, anyone looking to kickstart a career in Artificial Intelligence or looking upskill in this transformative field; this course provides the best opportunity.

Course Objectives:
•   To understand the fundamentals of Deep Learning, its techniques and applications in detail.
•   To make the participants well-versed in using various python libraries for Deep Learning.
•   To provide a solid foundation in neural network architectures, such as artificial neural networks (ANNs), convolutional neural networks (CNNs) and recurrent neural networks (RNNs) etc.
•   To make the participants comfortable in using deep neural networks to solve complex problems and predict outcomes from large datasets.
•   To make them able to use suitable deep learning algorithms and techniques for solving real-world problems.

Batch Details:
Class Timings: 7:00 pm – 9:00 pm (Wed & Friday) Start Date: 26th Mar 2025
Duration: 64 Hours End Date: 11th July 2025
Mode: Online Certification: iHUB Divyasampark IIT Roorkee
Last Date to Register: 25th Mar 2025

Course Fee: Students/PhD Scholars/RA/JRF/SRF/Postdoc fellows: Rs. 11,000/-
Faculty/Working Professional: Rs. 13,000/-
(Amounts inclusive of GST)

Course Highlights:
•   Industry-Relevant Skills.
• Hands-on based learning experience through practical projects.
• Globally accepted certification from iHUB Divyasampark IIT Roorkee
•   Full-time access to recorded lectures/PPTs/PDFs/Study Materials.
•   Session on Resume Preparation/Interview Preparation.

Course Overview:

Module 1

Basics of Python: Objects and Data Types.
Statements like if-elif-else, for, while etc., Methods and Functions.
Object Oriented Programming (OOPs) with Python.
NumPy Basics: Arrays, Index Selection and Operations.
Python for Data visualization & plotting using libraries like Matplotlib.
PyTorch & Tensor Basics and Operations for Deep Learning Applications.

Module 2

Machine Learning Basics, Introduction to supervised & unsupervised learning, Bias-Variance Trade-off, Loss Functions, Gradient Descent.
Linear Regression for One & Multiple Variables
Ordinary Least Square, Dummy Variables, One Hot Encoding, Polynomial Regression
Anscombe’s quartet, Performance Metrics like Mean Absolute Error, Root Mean Squared Error
Logistic Regression, Sigmoid Function, Anscombe’s quartet
Confusion Matrix, interpreting parameters like F-1 score, Accuracy, Precision, Recall etc.

Module 3

Introduction to Deep Learning
Artificial Neural Networks
Perceptron Model, Activation Functions
Forward and Backward Propagation
Bias-variance trade off, Overfitting & Underfitting of Deep Learning Models
Discussion on Keras and TensorFlow

Module 4

Image and Pixels basics for Neural Networks.
Convolutional Neural Networks, Pooling Layers, Convolutional Layers
Striding, Image Kernels and Filters, Convolution Kernels
CNNs for Computer Vision : Object Detection and Interpretation in Images.
Local connectivity, Pooling/Downsampling Layer.

Module 5

Max Pooling, Dropout Technique, AlexNet
Recurrent Neural Networks, Sequential Information, Recurrent Neurons, “Unrolled” layer.
Sequence to Sequence, Sequence to Vector, Vector to Sequence, Exploding Gradients, Vanishing Gradient
Long Short-Term Memory (LSTM) networks, RNN on a Sine Wave
Multivariate Time Series with LSTM RNN.

Module 6

Natural Language Processing with Deep Learning: Embedding Layer.
Gated Recurrent Unit, Dense
Text Vectorization and Encoding Dictionary
Autoencoders: Dimensionality Reduction and Noise Removal
Generative Adversarial Networks (GANs): Generators and Discriminators, Deep Convolutional GANs (DCGANs)
Capstone Project

Prerequisites and eligibility:

No coding experience in any programming language required. We’ll start from scratch.
This course can be taken up by any undergraduate/postgraduate student of Basic & Applied Sciences, Engineering, Management and Computer Applications and also by Research Scholars/Faculties/Working Professionals who want to upskill themselves.
Participants need to have a laptop/PC (with a minimum of 4 GB RAM, 100 GB HDD, Intel i3 processor) and proper internet/Wi-Fi connection.

Contact Person: Dr. Subrat Kotoky

Email: [email protected] / [email protected]

Phone: 9085317465 / 8473874389

Expert Profile: Mr. Shreyas Shukla

Professional Corporate Trainer & Microsoft Azure Certified Data Engineer

M.Tech-IIT Kharagpur & BE- The Aeronautical Society of India, New Delhi

Has successfully conducted 25+ courses and trained 1500+ learners in the fields of Python Programming, Data Analytics, Machine Learning, Deep Learning, Computer Vision etc. till now.

(Total Experience in conducting Professional Courses: 4+ Years)

Certifications: