Flow-based Deep Generative Models

📰 Lilian Weng's Blog

Flow-based generative models learn the probability density function of input data explicitly, differing from GAN and VAE

intermediate Published 13 Oct 2018
Action Steps
  1. Learn the basics of generative models, including GAN and VAE
  2. Understand the concept of probability density functions and their role in generative models
  3. Explore the architecture and training process of flow-based generative models
  4. Apply flow-based models to real-world data generation tasks
Who Needs to Know This

Data scientists and AI engineers can benefit from understanding flow-based models for generative tasks, as they provide an alternative approach to GAN and VAE

Key Insight

💡 Flow-based models provide an explicit and tractable way to learn probability density functions, differing from implicit models like GAN

Share This
🤖 Flow-based generative models explicitly learn probability density functions! 📊

Key Takeaways

Flow-based generative models learn the probability density function of input data explicitly, differing from GAN and VAE

Full Article

<!-- In this post, we are looking into the third type of generative models: flow-based generative models. Different from GAN and VAE, they explicitly learn the probability density function of the input data. --> <p>So far, I&rsquo;ve written about two types of generative models, <a href="https://lilianweng.github.io/posts/2017-08-20-gan/">GAN</a> and <a href="https://lilianweng.github.io/posts/2018-08-12-vae/">VAE</a>. Neither of them explicitly learns the probability density function of real da
Read full article → ← Back to Reads