What is Algorithmic Bias Mitigation
Techniques that help reduce unfair or systematic biases in data, models, and automated decisions.
Definition
Reducing bias doesn't start with the model, but with understanding the problem and the data. We need to examine which groups are underrepresented, which attributes may lead to discrimination, how errors are distributed, and which fairness criteria are important. Data cleaning, balancing, restrictions, auditing and monitoring are then applied.
Example
In the hiring model, the team checks whether the quality of recommendations for candidates from different regions, age groups or educational trajectories is degraded.
Why it matters
The term is important for responsible AI: a model may be accurate on average, but unfair for certain groups.
How it works
Approaches can be before training, during training and after it: improving data, changing the loss function, calibrating results and regular auditing.
Where it is used
- recruitment and HR
- credit scoring
- moderation and recommendations
Limitations
You cannot simply “remove a sensitive sign” and consider the problem solved: other signs can indirectly reproduce it.