Testing for bias · Question 4 of 60

What is meant by bias in an ML system?

1 answer is correct. Easy

1. A The time difference between the end of training and the start of testing
2. B The bias value each neuron adds to the weighted sum of its inputs before applying the activation function
3. C Random measurement errors in sensor data that cancel each other out on average
4. D Systematic, unjust unequal treatment because of traits such as gender or age