|
|
|
|
|
|
by bo1024
3222 days ago
|
|
|
> Math challenge: is there a way of combining the desirable properties mentioned in the post with the property of honesty? I suspect there is but I haven't tried it. The suggested method seems to be asking for binary responses (like/dislike), then aggregating them with the confidence-bound formula. This should be truthful in, e.g., a model where users who like it want to maximize the score and users who dislike it want to minimize the score. |
|
|