Last Week’s Work Review#
Option recognition
The choice of actions and policies based on natural language walkthrough is highly similar to “option recognition” in traditional AI study, where natural language walkthrough will help agents to make options for subpolicies selection.
When to start /end
When we want to let agent utilise walkthrough data, the agent needs to know when to execute the walkthrough.
- it needs to know what the current situation is and then know
- what we have already done
- where are we, what to execute next
- what to do next if the previous execution failed
- currently they are not explicitly handled
Reuse object recognition module
We can use go-explore’s object recognition module
You need password to access to the content, go to Slack *#phdsukai to find more.
Part of this article is encrypted with password:
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