[abrichr]

I'm having trouble fully grasping this:

> Today, rather than specify how to compute correct responses to requests, our engineers encode the means of providing feedback to a self-tuning system.

"encod[ing] the means of providing feedback to a self-tuning system", got it, very cool!

But don't they still have to "specify how to compute correct responses to requests"?

[nawgszy]

Not OP / GP; from what I understand, this isn't an API generated by ML, it's a cache manager. "Computing correct responses to requests" refers to deciding whether or not it should read it from some caching layer and whether or not it should be cached in the future, and it does this by optimizing some parameters. The difference is something like it decides whether or not to use the API or the cache to load an image or some content, rather than saying "this request should return a response with these properties". Hopefully I'm right and this makes sense.

[mohaps]

the basic difference in approach was to switch the earlier approach of

if (conditionA && conditionB && !conditionC) cache_it()

to

hey look, an item with featureset X is cacheable while one with featureset Y is not.

This reflects in the API which is just two calls predict() and feedback()

this simplifies the integration code and is easily debuggable even in the face of changes.