Search for: All records

Abstract Moving away from hazardous areas may be an important adaptive response under intensifying climate change, but to date such movement has been controversial and conducted with limited government or private-sector support. Research has emphasized resident perspectives on mobility, but understanding how professionals view it may open new avenues to shape future outcomes. Based on 76 interviews with professionals involved in climate responses in South Florida, we evaluate perceptions of adaptation goals, the potential role of climate mobilities in pathways supporting those goals, and associated constraints and enablers. The practitioners interviewed anticipate multiple types of climate mobilities will occur in the region, at increasing scales. Interviewees perceive climate mobilities at present, especially migration and gentrification where climate plays some role, as causing distributional inequities and financial and sociocultural disruptions, and they view existing adaptive strategies as best serving those who already have adequate resources, despite practitioners’ personal commitments to social justice goals. Although many practitioners feel prepared for their own, limited roles related to climate mobilities, they judge the region as a whole as being unprepared to support the retreat they see as inevitable, with a need for a more ambitious long-term transition plan. Achieving this need will be difficult, as practitioners indicate that climate mobilities remain hard to talk about politically. Nevertheless, interviewees believe some households are already considering moving in response to climate risks. Discussions of climate mobilities, through interviews and far beyond, may encourage more mindful choices about and engagement in climate-driven transformations. 

Many organizations seek to ensure that machine learning (ML) and artificial intelligence (AI) systems work as intended in production but currently do not have a cohesive methodology in place to do so. To fill this gap, we propose MLTE (Machine Learning Test and Evaluation, colloquially referred to as "melt"), a framework and implementation to evaluate ML models and systems. The framework compiles state-of-the-art evaluation techniques into an organizational process for interdisciplinary teams, including model developers, software engineers, system owners, and other stakeholders. MLTE tooling supports this process by providing a domain-specific language that teams can use to express model requirements, an infrastructure to define, generate, and collect ML evaluation metrics, and the means to communicate results.