Quantifying Algorithmic Biases over Time
This addresses the issue of understanding dynamic biases in algorithms for researchers and practitioners, though it is incremental as it extends static bias metrics to a temporal context.
The paper tackles the problem of measuring algorithmic biases that change over time, proposing new metrics and applying them to a case study of gender representation in Twitter image searches for #Nurse and #Doctor over 21 days, finding that biases vary significantly and can differ in direction daily.
Algorithms now permeate multiple aspects of human lives and multiple recent results have reported that these algorithms may have biases pertaining to gender, race, and other demographic characteristics. The metrics used to quantify such biases have still focused on a static notion of algorithms. However, algorithms evolve over time. For instance, Tay (a conversational bot launched by Microsoft) was arguably not biased at its launch but quickly became biased, sexist, and racist over time. We suggest a set of intuitive metrics to study the variations in biases over time and present the results for a case study for genders represented in images resulting from a Twitter image search for #Nurse and #Doctor over a period of 21 days. Results indicate that biases vary significantly over time and the direction of bias could appear to be different on different days. Hence, one-shot measurements may not suffice for understanding algorithmic bias, thus motivating further work on studying biases in algorithms over time.