You stuck that reward right in front of Rocky’s nose and he didn’t know what it meant at all. You’ve blinded him, all his trajectory calculations look the same! Didn’t even eat it. That’s one flat and lost derivative right there! It would be nice if you could give Rocky a treat for some incremental progress but you chose a Sparse Rewards training scheme and you’re just gonna have to wait this one out?
You Try to give Rocky a treat but he’s super confused
