IRL 5/​8: Maximum Causal Entropy IRL

Link post

Every Mon­day for 8 weeks, we will be post­ing les­sons about In­verse Re­in­force­ment Learn­ing. This is les­son 5. We are pub­lish­ing it now be­cause it would have in­terfered with the ran­dom­ized con­trol­led trial we are run­ning, and yes­ter­day we finished col­lect­ing re­sponses from the par­ti­ci­pants. A LW post with the re­sults will ap­pear in a few days. Fu­ture IRL les­sons will re­sume nor­mally on Mon­day.

Note that ac­cess to the les­sons re­quires cre­at­ing an ac­count here.

This les­son comes with the fol­low­ing sup­ple­men­tary ma­te­rial:

Have a nice day!