Deep Reinforcement Learning Hands-On

Chapter 13. Web Navigation

This chapter takes a look at another practical application of Reinforcement Learning (RL): web navigation and browser automation. We'll discuss why web navigation is important and how it can be solved with an RL approach. Next, we will take a deep look at one very interesting, but a commonly overlooked and a bit of abandoned RL benchmark, which was implemented by OpenAI and called Mini World of Bits.

Web navigation

When the web was invented, it started as several text-only web pages interconnected by hyperlinks. If you're curious, here is the first web page home: http://info.cern.ch/, with text and links. The only thing you can do is to read and click on links to go between pages. Several years later, in 1995, IETF published ...