Abstract: The term ‘Human in the Loop’ isn’t new, but many data scientists and analytics leaders don’t know what it is or why it has become so important. Surveys and industry interviews have revealed that it is not yet a focus of data science education, yet certain kinds of machine learning model development are nearly impossible without it. Specifically, it is the management, and even the ethics, of human-in-the-loop projects that many are unprepared for.
This 30-minute business and strategy-oriented talk aims to educate data scientists and analytics leaders about the role and benefits of human-in-the-loop (HITL), presented from the standpoint of a 25+ year veteran of traditional machine learning applications who has recently been immersed in this new world. The session will include brief examples in application areas as diverse as Medical AI, Ag Tech, and autonomous vehicles.
Lesson 1: Why is Deep Learning driving demand for Human in the Loop? The data science and machine learning community is still adjusting to the revolutionary charge triggered by the 2012 ImageNet competition. We will discuss how that same event has created a cottage industry helping to produce the very large datasets necessary to contemporary machine learning.
Lesson 2: HITL isn't just about training data
We'll discuss that human-in-the-loop can be a powerful strategic complement to machine learning models during model development. HITL is primarily associated with creating large training datasets, but it can also be helpful in exception processing.
Lesson 3: The education gap: Data Science education and HITL
We will discuss some possible reasons that relatively few data scientists are familiar with managing human-in-the-loop projects. We will also identify some of the key considerations that practitioners and analytics leaders should know.
Lesson 4: The emerging application areas for HITL
Wherever you find deep learning, you also find the need for human-in-the-loop. We'll explore some of the most compelling application areas, including Medical AI, Ag Tech, and drone delivery.
General reference will be made to modeling techniques like traditional machine learning and deep learning, but there are no specific prerequisites.
Bio: Keith McCormick serves as CloudFactory’s Chief Data Science Advisor. He's also an author, LinkedIn Learning contributor, university instructor, and conference speaker. Keith has been building predictive analytics models since the late 90s. More recently his focus has shifted to helping organizations build and manage their data science teams.