[question] adanet with Reinforcement Learning?

[jarlva]

RL is very time and resource intensive, often taking hours and days to complete. Any reduction/optimization is a huge benefit.
Has anyone tried to apply adanet with RL?

[cweill]

@jheffez Our team has mostly looked into applying adanet to supervised learning tasks, and some semi-supervised learning work.

If you think there is a good way to use adanet for RL, I recommend sharing a notebook or writing a paper about it :).