The ethical concept of fairness has recently been applied in machine learning (ML) settings to describe a wide range of constraints and objectives. When considering the relevance of ethical concepts to subset selection problems, the concepts of diversity and inclusion are additionally applicable in order to create outputs that account for social power and access differentials. We introduce metrics based on these concepts, which can be applied together, separately, and in tandem with additional fairness constraints. Results from human subject experiments lend support to the proposed criteria. Social choice methods can additionally be leveraged to aggregate and choose preferable sets, and we detail how these may be applied. CCS CONCEPTS• Information systems → Information retrieval diversity; Evaluation of retrieval results. KEYWORDS machine learning fairness, subset selection, diversity and inclusion ACM Reference Format:

Online social media platforms increasingly rely on Natural Language Processing (NLP) techniques to detect abusive content at scale in order to mitigate the harms it causes to their users. However, these techniques suffer from various sampling and association biases present in training data, often resulting in sub-par performance on content relevant to marginalized groups, potentially furthering disproportionate harms towards them. Studies on such biases so far have focused on only a handful of axes of disparities and subgroups that have annotations/lexicons available. Consequently, biases concerning non-Western contexts are largely ignored in the literature. In this paper, we introduce a weakly supervised method to robustly detect lexical biases in broader geocultural contexts. Through a case study on a publicly available toxicity detection model, we demonstrate that our method identifies salient groups of cross-geographic errors, and, in a follow up, demonstrate that these groupings reflect human judgments of offensive and inoffensive language in those geographic contexts. We also conduct analysis of a model trained on a dataset with ground truth labels to better understand these biases, and present preliminary mitigation experiments.

No abstract

No abstract

The Hikurangi Margin east of New Zealand’s North Island hosts an extensive gas hydrate province with numerous gas hydrate accumulations related to the faulted structure of the accretionary wedge. One such hydrate feature occurs in a small perched upper-slope basin known as Urutī Basin. We investigate this hydrate accumulation by combining a long-offset seismic line (10-km-long receiver array) with a grid of high-resolution seismic lines acquired with a 600-m-long hydrophone streamer. The long-offset data enable quantitative velocity analysis while the high-resolution data constrain the three-dimensional geometry of the hydrate accumulation. The sediments in Urutī Basin dip landward due to ongoing deformation of the accretionary wedge. These strata are clearly imaged in seismic data where they cross a distinct bottom simulating reflection (BSR) that dips, counterintuitively, in the opposite direction to the regional dip of the seafloor. BSR-derived heat flow estimates reveal a distinct heat flow anomaly that coincides spatially with the upper extent of a landward-verging thrust fault. We present a conceptual model of this gas hydrate system that highlights the roles of fault-controlled fluid flow at depth merging into strata-controlled fluid flow into the hydrate stability zone. The result is a layer-constrained accumulation of concentrated gas hydrate in the dipping strata. Our study provides new insight into the interplay between deep faulting, fluid flow and the shallow processes involved in gas hydrate formation.