Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!

Grab it

(This program was formerly part of a three-course specialization called Autonomous AI for Industry. Because the software program Bonsai was discontinued, references to Bonsai have been removed. You can still learn about autonomous AI and machine teaching through our two individual courses "Designing Autonomous AI" and "Machine Teaching for Autonomous AI.") To design an autonomous AI system, you must figure out how to distill a business challenge into its component parts. When children learn how to hit a baseball, they don’t start with fastballs. Their coaches begin with the basics: how to grip the handle of the bat, where to put their feet and how to keep their eyes on the ball. Similarly, an autonomous AI system needs a subject matter expert (SME) to break a complex process or problem into easier tasks that give the AI important clues about how to find a solution faster. In this course, you’ll learn how to create an autonomous AI design plan. By setting goals, identifying trainable skills, and employing those skills in goal-oriented strategies, you’ll incorporate your SME’s knowledge directly into your AI’s “brain,” the agent that powers your autonomous system. You'll learn when and how to combine various AI architecture design patterns, as well as how to design an advanced AI at the architectural level without worrying about the implementation of neural networks or machine learning algorithms. At the end of this course, you’ll be able to: • Interview SMEs to extract their unique knowledge about a system or process • Combine reinforcement learning with expert rules, optimization and mathematical calculations in an AI brain • Design an autonomous AI brain from modular components to guide the learning process for a particular task • Validate your brain design against existing expertise and techniques for solving problems • Produce a detailed specifications document so that someone else can build your AI brain

Syllabus

Defining your AI

The first step in designing an autonomous AI is defining what your AI is going to do and what the goals are. Think about it like describing a game to someone. First you explain what the object of the game is, and then you describe the rules. In this module you'll learn how to do the same for your autonomous AI use case.

Teaching Skills to your AI

Autonomous AI brains are built from skills. Skills are “units of competence for completing tasks that have sub-goals associated with them.” This week you'll learn to outline the skills you want your autonomous AI brain to learn. First, you’ll identify three different types of skills that you can build into your brains. Then, you’ll learn a strategy that will help easily extract and document skills from subject matter experts you interview. Along the way we’ll look at some design patterns that you can use as templates to start your use case brain designs.

Organizing Skills in your AI

Now that you understand how to interview a subject matter expert and lay out all the skills that you want your AI to practice, you need to organize those skills in the brain. In this module you’ll learn two organizing paradigms for skills in autonomous AI, and a three-step framework for completing this orchestration. This week you’ll see some brain design patterns for example use cases, to help you with thinking about organizing your own use case brain design.

Putting it All Together

Now it's time to put it all together. You've defined your AI, you've identified a set of skills that you want to teach your AI and you've used brain design patterns in the paradigms of orchestration to snap those skills together in the right arrangement. There's a few pitfalls to orchestration that you should be aware of and you’ll have lots of opportunity to practice creating variations on brain designs for sample problems. Make sure to share your brain designs from the lab in the forum, so we can discuss them together and learn from each other.