IIMBx offers two new courses

IIM Bangalore’s digital learning vertical, IIMBx, has added two more courses to its portfolio, ‘Reinforcement Learning Algorithms’ and ‘Artificial Neural Networks and Deep Learning’. 

Reinforcement Learning Algorithms

IIMBx has developed a new course on algorithms that will help one make decisions under uncertainty. The course, titled, ‘Reinforcement Learning Algorithms’, will be released as a part of IIMBx’s ongoing programme: ‘Artificial Intelligence for Managers’. 

Click here for details: https://iimbx.iimb.ac.in/ai-for-managers/

Reinforcement Learning Algorithms is a five-week course. It deals with problems in which an agent needs to make a sequence of decisions under uncertainty, and the objective is to maximize the reward earned through each decision. 

RL offers a powerful analytical approach to model and examines complex finance, retail, marketing, operations and economics problems under uncertainty. In management as well as in business, many measurements change with time and are inherently random in nature. This module introduces stochastic processes and their applications to business and management. Stochastic models are also the basis for reinforcement learning algorithms.

This course will help learners get introduced to stochastic models, Markov models and other tools that will help them use algorithms for decision-making. Participants will learn the concepts of Classification of States, Steady-state Probability Estimation, Brand Switching and Loyalty Modelling, Market Share Estimation in the short and long run, Google’s Ranking Algorithm, and use of the Poisson Process in Operations, Marketing and Insurance. 

This new course will help learners apply the Reinforcement Learning algorithm techniques for Dynamic Programming, Markov Decision Processes in Sequential Decision-making, Policy Iteration and Value Iteration Algorithm. The course has live sessions on Harvard Business Publishing (HBP) cases, discussions and recorded lecture videos. 

Artificial Neural Networks and Deep Learning 

IIMBx has also launched another new course titled, ‘Artificial Neural Networks and Deep Learning’, as part of its programme – Artificial Intelligence for Managers. 

Click here for details: https://iimbx.iimb.ac.in/ai-for-managers/

This is an eight-week course, which will provide a strong foundation in Deep Learning using TensorFlow/Keras, by providing real-life case studies and examples. The course begins with the concept of representational learning, understanding the difference between machine learning and deep learning, and listing the factors leading to deep learning’s prominence. 

The course will gradually dive deeper into artificial neural networks by covering topics ranging from building and training simple neurons, perceptron, etc., using algorithms such as gradient descent, back-propagation, etc., to deep network/architecture with hyper-parameter tuning.

Deep learning has solved complex problems in computer vision, natural language processing, etc., through diverse architecture such as convolutional neural networks, recurrent neural networks, transformer models, etc. The course will cover all this architecture in detail and provide a step-by-step approach to exploring and building these models.

The course will help participants learn: 

• Artificial Neural Networks: Biological and Mathematical Neurons; Perceptron Model – Multi-Layer Perceptron; Back-propagation Algorithm; Activation Functions
• Introduction to Artificial Intelligence and Deep Learning: Relationship between Machine Learning and Deep Learning
• Learning Process: Representational Learning
• Deep Learning’s Application Area; Challenges and Frameworks
• Training Deep Neural Networks: Hyper-parameter Tuning, Optimizers, Addressing Overfitting Issues: Regularization and Dropout
• Unsupervized Learning using Autoencoders and Transfer Learning
• Convolutional Neural Network (CNN) and Computer Vision

The course will help participants understand the concept of convolutions and various operations/layers, such as, padding, striding, pooling, etc., which form convolutional neural networks.

It will also throw light on building and training CNNs for computer vision problems, such as image classification.

In detail, participants will understand CNN architecture such as VGG, ResNet, Inception, Xception, etc., and gain knowledge on how to perform transfer learning by applying these pre-trained models and weights on different datasets.