search

facebookresearch / ReAgent

A platform for Reasoning systems (Reinforcement Learning, Contextual Bandits, etc.)
https://reagent.ai
BSD 3-Clause "New" or "Revised" License
3.58k stars 521 forks source link

SlateQ agent [Q & A] #699

Open getsanjeevdubey opened 2 years ago

getsanjeevdubey commented 2 years ago

Hi @kittipatv Do we have any resource to understand SlateQ algorithm implemented here; apart from original paper. I am exploring this for a recommendation problem but have some doubts around it.

getsanjeevdubey commented 2 years ago

Particularly:

  1. Does SlateQ agent assume that user is going to click at-least one of the items? What if user does not click any of items?
  2. What is the role of no_click action?
  3. Q-value of no_click action signifies what?

Any response would be appreciated.