Knowing about algorithm bias can help us steer clear of a future where Artificial Intelligence are used harmful, discriminatory ways.
1. Data reflects existing biases
2. Unbalanced classes in training data
3. Data does not capture the right value
4. Data amplified by a feedback loop
A possible feedback loop basically means amplifying what happened in the past. (PredPol)
5. Malicious data at lack or manipulation
The first step is just understanding that algorithms will be biased. It's mportant to be critical about Al recommendations, instead of
just accepting that the computer said so.
Second, if we want to have less biased algorithms, we may need more training data on protected classes like race, gender, or age.
Looking at an algorithm's recommendations for protected classes may be a good way to check it for discrimination.
评论