3 ms·
Not end-to-end learning. They've included a lot of apriori information into their models to significantly constrain the search space. Maybe I'm being a little t
by rayuela 10y ago
Not end-to-end learning. They've included a lot of apriori information into their models to significantly constrain the search space. Maybe I'm being a little too harsh in my opinions, but with this work coming out of CMU I was expecting a little more.
- shmel 10y agoHowever it runs on test time from raw pixels only.
- devindotcom 10y agoThey trained it with some flags to say whether an enemy or item was onscreen, but it ran on pixel data only. I'm sure a bit more work could get it to train on pixels only as well, but my guess is it would take more time and tweaking than they were capable of applying.