Sai Krishna Gottipati
TalkRL: The Reinforcement Learning Podcast
00:00
The Relationship Between Paird and the Pard System
The work can be seen as a specific instance of the broader problem of this gent echo systems, where you have one agent generating increasingly complex environments. All those al thems that are applicable in those imagined eco systems are applicable here as well. So we have tried this approach on two environments. One is fetch, and the other is a custom driving similator. In both he cases, we represent the position as x, y and yer. We could try other things, for example, as a bit map representation if it's a rebolt kind of setting.
Play episode from 31:12
Transcript
