About this course
This comprehensive course introduces learners to the world of generative artificial intelligence (AI) models within machine learning (ML), focusing on Amazon SageMaker as a prime tool for design, training, optimization, deployment, and monitoring. In this course, we will begin with a deep dive into the fundamental concepts and types of generative models like generative adversarial networks (GANs) and variational autoencoders (VAEs). We will explore their pros, cons, and relevant architectures. Next, we will use hands-on tutorials and in-depth discussions to guide you through the steps of designing a GAN architecture, leveraging SageMaker’s built-in algorithms, preprocessing data, and distributed training capabilities. We will then delve into optimization techniques, transfer learning, and quality evaluation methods, looking at ways apply these concepts in real-world scenarios. Lastly, we will introduce deployment strategies in SageMaker, highlighting how to serve models as endpoints for real-time inferences and how to efficiently monitor and troubleshoot their generative models.
Learning objectives
Discover the key concepts covered in this course
Outline the fundamental concepts of generative models and their applications in machine learning
Analyze the different types of generative models, including generative adversarial networks (gans) and variational autoencoders (vaes)
Show all