It sounds exactly like A/B testing, but using a specific algorithm to determine the winner.
It talks about comparing the current situation to the best situation... But in most A/B testing, A would be the current 'best' and B would be the challenger. Same thing.
It also talks about a reward for certain button-presses, but that isn't actually what you want to optimize. You want to optimize revenue. So it's possible this could send you down the wrong path.
And if it's saying you should pit the current site against the best the site has done historically, that's ridiculous. You couldn't possibly put controls on all of the factors involved. That's why A/B testing is special: All the other factors are guaranteed to be as identical as possible.
No, it's not A/B testing for the reasons I try to explain the in the post. A/B testing can't change the choices while the experiment is running, doesn't adjust to customer preferences in real-time, etc.
You can optimise for revenue. Button presses was just a simple example.
Can you compare to MVT (Multi-Variate Testing)? From an admittedly surface reading of the post, this sounds like the somewhat common "A/B sucks, MVT is more accurate at optimizing and showing impact" with a newer optimization approach.
In MVT you're interesting in testing the interactions between different elements. If the elements are encountered in sequence the bandit analogue is reinforcement learning. If you have one page with different "slots" to fill then "bandit slate" algorithms might be appropriate. The key advantage is that all these approaches are online: they take advantage of information as it is received, and you can change things and the algorithms adapt. A/B testing and MVT
don't do either.
It talks about comparing the current situation to the best situation... But in most A/B testing, A would be the current 'best' and B would be the challenger. Same thing.
It also talks about a reward for certain button-presses, but that isn't actually what you want to optimize. You want to optimize revenue. So it's possible this could send you down the wrong path.
And if it's saying you should pit the current site against the best the site has done historically, that's ridiculous. You couldn't possibly put controls on all of the factors involved. That's why A/B testing is special: All the other factors are guaranteed to be as identical as possible.