The observed behavior is the same: using the new model, most people are still shown highly polarized posts, as indicated by subjective assessment of user research professionals.
What should you do now?
The observed behavior is the same: using the new model, most people are still shown highly polarized posts, as indicated by subjective assessment of user research professionals.
What should you do now?
This is because of the fact that machine learning models are prone to learn quite different things than the objective function intended, hence the introduction of different intent or structure of the model must be disregarded when analysing the results.
To any degree the models predict similarly, they must be regarded as similar, but perhaps in a roundabout way.
* The 'engagement' metric leads to toxic outcomes no mater what.
* The upper management / board is single mindedly obsessed with 'engagement', as a proxy for making money.
* I cannot function in an environment where my personal ethics is in direct conflict with the company focus.
Therefore I quit. YMMV.