PG Program in Advanced Artificial Intelligence for Business Applications

This course will help you go deep into the workings of neural networks. Rather than the Neural Network process being a black box, you will understand what drives performance, and be able to more systematically get good results. You will also learn about some of the advanced optimization techniques.

Autoencoders are a very popular neural network architecture in Deep Learning. Here you will dive in the realms of applying advanced methods on the problems in the area of data augmentation and data encoding for efficient compression and modeling of data for meaningful purposes in the fields of computer vision, computer networks, computer architecture, and many others.

In this course you will learn about how CNN learns to do feature extraction and the core concept of CNN. You will learn how to build convolutional neural networks and apply them to image data, and about the popular CNN architectures like AlexNes, GoogleNet, ResNet, ExciationNet etc. Going one step further we also learn more advanced techniques - Transfer Learning.

We will learn about how to use variations of CNN to different object recognition tasks. Object detection is applied in numerous territories of image processing, including picture retrieval, security, observation, computerized vehicle systems and machine investigation.

Learn about the essential image analysis task - Semantic Segmentation. There are cutting edge applications like self-driving vehicles, human-computer interaction, virtual reality etc. Semantic segmentation is one of the high-level tasks that paves the way towards complete scene understanding.

This course will enable you to explore the working behind numerous exciting applications of accurate face recognition using CNNs. There are many useful applications of this, in the area of arts generation, social media engagement, security and access control etc. You’ll learn to implement your own algorithm to generate art and recognize faces!

Explainable AI (XAI) is a set of tools and frameworks to help you understand and interpret predictions made by your machine learning models.

POS tagging finds applications in Named Entity Recognition (NER), sentiment analysis, question answering, and word sense disambiguation. We will look at an example of word sense disambiguation. Use of Named Entity recognition and Part of Speech tagging in NLP. Representation of Words in vector space and its visualization, Word2vec, word embeddings, use of these techniques.

Learn how to work with natural language data after deep dive into the world of language models of Sequential Models and state of the art language models. Use of Named Entity recognition and Part of Speech tagging in NLP.

Long Short Term Memory networks are really useful to classify and predict on sequential data.

Going one step further you will learn about advanced Sequential models, Transformer Networks (aims to solve sequence-to-sequence tasks while handling long-range dependencies), and attention models. Both sequence-to-sequence tasks, like Machine Translation, Document Summarization, etc. and classification tasks, like Sentiment Analysis, PoS Tagging, Document Classification, etc. benefit by use of attention models..

Deployment or pushing AI models into production means making your models available to the end users or systems. This course is aimed to provide a glimpse into one of the key activities of a "Full-Stack Data Scientist" which involves deployment of a model into a production environment.

The knowledge of deploying the models shall be handy when you join startups or work with cross-functional teams in your organisation.

Starting with the motive behind using the terms in GANs, we look at the individual constituents of a Generative Adversarial Network. Building familiarity with the Generator and Discriminator, we will solve some problems using them.

Various applications of GAN are in digital image processing, medical imaging analysis, as well as the latest applications of GAN in medical informatics and bioinformatics, Cybersecurity.

We learn what is Reinforcement Learning, its applications, key approaches to building a RL framework and various processes involved in it.

State–action–reward–state–action (SARSA) is an algorithm for learning a Markov decision process policy, used in the reinforcement learning area of machine learning.

We learn what is Reinforcement Learning, its applications, key approaches to building a RL framework and various processes involved in it. There are various applications of Reinforcement learning in Robotics for industrial automation, Business strategy planning, and creating training systems that provide custom instruction and materials according to the requirement of students.