Re:YouTube’s Algorithm Incentivizes the Wrong Behavior
by dbt00 on Friday June 14, 2019 @04:46PM
(#20186545)
The problem of the dataset is not just that AI will pick shortcuts and naive heuristics, because humans will too.
The problem of the dataset is that you're not in control of who populates the dataset and what their intentions are. There's no understanding of an adversarial model and threat handling.
The problem of the dataset is that you're not in control of who populates the dataset and what their intentions are. There's no understanding of an adversarial model and threat handling.