Search for: All records

As reliance on Machine Learning (ML) systems in real-world decision-making processes grows, ensuring these systems are free of bias against sensitive demographic groups is of increasing importance. Existing techniques for automatically debiasing ML models generally require access to either the models’ internal architectures, the models’ training datasets, or both. In this paper we outline the reasons why such requirements are disadvantageous, and present an alternative novel debiasing system that is both data- and model-agnostic. We implement this system as a Reinforcement Learning Agent and through extensive experiments show that we can debias a variety of target ML model architectures over three benchmark datasets. Our results show performance comparable to data- and/or model-gnostic state-of-the-art debiasers. 

Abstract Decadal scale lake drying in interior Alaska results in lake margin colonization by willow shrub and graminoid vegetation, but the effects of these changes on plant production, biodiversity, soil properties, and soil microbial communities are not well known. We studied changes in soil organic carbon (SOC) and nitrogen (N) storage, plant and microbial community composition, and soil microbial activities in drying and non‐drying lakes in the Yukon Flats National Wildlife Refuge. Historic changes in lake area were determined using Landsat imagery. Results showed that SOC storage in drying lake margins declined by 0.13 kg C m⁻² yr⁻¹over 30 years of exposure of lake sediments, with no significant change in soil N. Lake drying resulted in an increase in graminoid and shrub aboveground net primary production (ANPP, +3% yr⁻¹) with little change in plant functional composition. Increases in ANPP were similar in magnitude (but opposite in sign) to losses in SOC over a 30‐year drying trend. Potential decomposition rates and soil enzyme activities were lower in drying lake margins compared to stable lake margins, possibly due to high salinities in drying lake margin soils. Microbial communities shifted in response to changing plant communities, although they still retained a legacy of the previous plant community. Understanding how changing lake hydrology impacts the ecology and biogeochemistry of lake margin terrestrial ecosystems is an underexamined phenomenon with large impacts to landscape processes. 

Urban population growth has significantly complicated the management of mobility systems, demanding innovative tools for planning. Generative Crowd-Flow  (GCF) models, which leverage machine learning to simulate urban movement patterns, offer a promising solution but lack sufficient evaluation of their fairness–a critical factor for equitable urban planning. We present an approach to measure and benchmark the fairness of GCF  models by developing a first-of-its-kind set of fairness metrics specifically tailored for this purpose. Using observed flow data, we employ a stochastic biased sampling approach to generate multiple permutations of Origin-Destination  datasets, each demonstrating intentional bias. Our proposed framework allows for the comparison of multiple GCF  models to evaluate how models introduce bias in outputs. Preliminary results indicate a tradeoff between model accuracy and fairness, underscoring the need for careful consideration in the deployment of these technologies. To this end, this study bridges the gap between human mobility literature and fairness in machine learning, with potential to help urban planners and policymakers leverage GCF  models for more equitable urban infrastructure development. 

Abstract Permafrost underlies approximately one quarter of Northern Hemisphere terrestrial surfaces and contains 25–50% of the global soil carbon (C) pool. Permafrost soils and the C stocks within are vulnerable to ongoing and future projected climate warming. The biogeography of microbial communities inhabiting permafrost has not been examined beyond a small number of sites focused on local-scale variation. Permafrost is different from other soils. Perennially frozen conditions in permafrost dictate that microbial communities do not turn over quickly, thus possibly providing strong linkages to past environments. Thus, the factors structuring the composition and function of microbial communities may differ from patterns observed in other terrestrial environments. Here, we analyzed 133 permafrost metagenomes from North America, Europe, and Asia. Permafrost biodiversity and taxonomic distribution varied in relation to pH, latitude and soil depth. The distribution of genes differed by latitude, soil depth, age, and pH. Genes that were the most highly variable across all sites were associated with energy metabolism and C-assimilation. Specifically, methanogenesis, fermentation, nitrate reduction, and replenishment of citric acid cycle intermediates. This suggests that adaptations to energy acquisition and substrate availability are among some of the strongest selective pressures shaping permafrost microbial communities. The spatial variation in metabolic potential has primed communities for specific biogeochemical processes as soils thaw due to climate change, which could cause regional- to global- scale variation in C and nitrogen processing and greenhouse gas emissions. 

Because of its high phosphorus (P) demands, it is likely that the abundance, distribution, and N-fixing capacity of Alnus in boreal forests are tightly coupled with P availability and the mobilization and uptake of soil P via ectomycorrhizal fungi (EMF). We examined whether Alnus shifts EMF communities in coordination with increasingly more complex organic P forms across a 200-year-old successional sequence along the Tanana River in interior Alaska. Root-tip activities of acid phosphatase, phosphodiesterase, and phytase of A. tenuifolia-associated EMF were positively intercorrelated but did not change in a predictable manner across the shrub, to hardwood to coniferous forest successional sequence. Approximately half of all Alnus roots were colonized by Alnicola and Tomentella taxa, and ordination analysis indicated that the EMF community on Alnus is a relatively distinct, host-specific group. Despite differences in the activities of the two Alnus dominants to mobilize acid phosphatase and phosphodiesterase, the root-tip activities of P-mobilizing enzymes of the Alnus-EMF community were not dramatically different from other co-occurring boreal plant hosts. This suggests that if Alnus has a greater influence on P cycling than other plant functional types, additional factors influencing P mobilization and uptake at the root and/or whole-plant level must be involved.