3 credits | core course

An introduction to basic concepts, methodology, algorithms, and technology used in business analytics and decision making. Explore concepts from probabilistic modeling, analysis and experimental design. Examine the basics of modern regression and classification, clustering, visualization, dimensionality reduction, A/B Testing and an introduction to deep learning.

1 credit | elective course

This course is a survey of some of the emerging machine learning: (i) Models are of increasingly larger size (e.g., foundation models such as BERT and GPT), but are trained mostly in an unsupervised manner (contrastive learning + supervised fine-tuning); (ii) Training of models are often distributed due to data privacy (e.g., in healthcare or finance); and (iii) Beyond inference (classification/regression), models are being used for decision-making (bandits and reinforcement learning). Applications of these ideas are important in a variety of settings, including recommendation systems, large language models, personalized models, online advertising, A/B testing and autonomous driving.

 This class will discuss some of the learning approaches that underlie these trends. We will discuss contrastive learning and model fine-tuning for (mostly) unsupervised model training (used in computer vision, large language models), and federated learning for distributed model training (used in healthcare, finance). We will then cover bandits and reinforcement learning for decision-making (used in healthcare, online advertising and recommendations, autonomous driving). We will overview the basic probability concepts, and introduce algorithmic approaches to offline and online decision-making (overview of bandits and reinforcement learning). The class will be structured around a sequence of programming assignments that will provide intuition on the application of these ideas.

Digital Technologies and Business Innovations