Quantifying Program Bias
This addresses the need for automated verification of bias in algorithms, which is critical for fairness in sensitive applications, representing a novel method for a known bottleneck.
The paper tackles the problem of quantifying bias in decision-making programs by proposing a novel probabilistic program analysis technique, resulting in the development of FairSquare, the first verification tool for this purpose, which was evaluated on various programs.
With the range and sensitivity of algorithmic decisions expanding at a break-neck speed, it is imperative that we aggressively investigate whether programs are biased. We propose a novel probabilistic program analysis technique and apply it to quantifying bias in decision-making programs. Specifically, we (i) present a sound and complete automated verification technique for proving quantitative properties of probabilistic programs; (ii) show that certain notions of bias, recently proposed in the fairness literature, can be phrased as quantitative correctness properties; and (iii) present FairSquare, the first verification tool for quantifying program bias, and evaluate it on a range of decision-making programs.