Share Description
You will learn how to assess the commercial value of an AI product. You'll begin by becoming acquainted with and fluent in common AI concepts. After that, you'll learn how to scope and build a data set, train a model, and assess its business impact. Finally, you'll discover how to ensure the success of a product by focusing on scalability, potential biases, and compliance.
Syllabus:
Course 1: Introduction to AI in Business
Introduction to AI and Machine Learning
- Learn the basics of AI and machine learning, and how businesses derive value from AI
- Understand the meaning of key terminologies, such as learning, unsupervised learning, and neural networks
Using AI and ML in Business
- Learn to narrow down a business use case and decide when to use AI in a product
- Learn strategies for measuring the success of a product
- See how to build an AI product team that can manage data and test product efficacy, over time
Course 2: Create a Dataset
Data Fit & Annotation
- Learn to analyze the size of your data and how well data fits a particular product use case
- Learn how to use Figure Eight’s crowdsourced data annotation platform to generate a high-quality ground-truth dataset with human annotation
- Design annotation instructions for best-in-class results
Project: Medical Image Annotation
- Define a product goal for a medical diagnostic tool
- Design an annotation job for a medical image dataset
- Consider metrics for success, how you might improve the annotation design, and design test questions for annotators
Project: Create a Medical Image Annotation Job
Learn how to design a data annotation job to generate a new dataset. As a product owner in this medical image annotation project, your goal is to create a product that assists doctors in quickly identifying cases of pneumonia in children. You'll want to create a classification system that can help doctors identify serious cases of pneumonia and act as a diagnostic aid. Your main task will be to create a data labeling job using Figure Eight's platform, which can be found at https://www.figure-eight.com/platform/.
Course 3: Build a Model
Training and Evaluating a Model
- Learn how a neural network learns from training data.
- Use test data to evaluate a trained model according to metrics like accuracy, precision, and recall.
- Learn how to use pre-trained models to transfer learning from one resource to another.
Project: Build a Model
- Build and train a model using Google’s AutoML.
- Evaluate several models and decide on the best model for a given product use case.
Project: Build a Model with Google AutoML
Learn how to build models using automated ML in this project, from data to results (no coding required). Create your own model for a medical imaging use case using Google AutoML. Then, run the model with four different variants of the data to see how the data affects the model's performance.
Course 4: Measuring Impact and Updating Models
Measuring Business Impact & Mitigating Bias
- Learn how to measure the business outcomes of a launched product.
- Discuss A/B testing and versioning.
- Learn strategies for mitigating unwanted bias in a machine learning model and product.
- See how to scale a product, according to user audience and demand.
Case Study: Video Annotation
- See an end-to-end AI product development cycle.
- Learn strategies for ideating solutions to problems, and improving a prototype, incrementally.
- Spend your time focused on prototyping a product, and learn strategies for continuously learning and updating a machine learning model.
Project: Capstone Proposal
- Develop a business proposal for an AI product.
Project: Capstone Proposal
You will create a business proposal for an AI product based on a use case of your choice for the capstone project. You will create a product business case, define success metrics, scope the dataset, plan model development, and create a post-deployment monitoring plan. Reviewers will assess the rigour and completeness of your proposal.
