Search

While advances in mobility technology including autonomous vehicles and multi-modal navigation systems can improve mobility equity for people with disabilities, these technologies depend crucially on accurate, standardized, and complete pedestrian path networks. Ad hoc collection efforts lead to a data record that is sparse, unreliable, and non-interoperable. This paper presents a sociotechnical methodology to collect, manage, serve, and maintain pedestrian path data at a statewide scale. Combining the automation afforded by computer-vision approaches applied to aerial imagery and existing road network data with the quality control afforded by interactive tools, we aim to produce routable pedestrian pathways for the entire State of Washington within approximately two years. We extract paths, crossings, and curb ramps at scale from aerial imagery, integrating multi-input segmentation methods with road topology data to ensure connected, routable networks. We then organize the predictions into project regions selected for their value to the public interest, where each project region is divided into intersection-scale tasks. These tasks are assigned and tracked through an interactive tool that manages concurrency, progress, feedback, and data management. We demonstrate that our automated systems outperform state-of-the-art methods in producing routable pathway networks, which then significantly reduces the time required for human vetting. Our results demonstrate the feasibility of yielding accurate, robust pedestrian pathway networks at the scale of an entire state. This paper intends to inform procedures for national-scale ADA compliance by providing pedestrian equity, safety, and accessibility, and improving urban environments for all users., Comment: arXiv admin note: text overlap with arXiv:2303.02323

This research introduces the Multilevel Embedding Association Test (ML-EAT), a method designed for interpretable and transparent measurement of intrinsic bias in language technologies. The ML-EAT addresses issues of ambiguity and difficulty in interpreting the traditional EAT measurement by quantifying bias at three levels of increasing granularity: the differential association between two target concepts with two attribute concepts; the individual effect size of each target concept with two attribute concepts; and the association between each individual target concept and each individual attribute concept. Using the ML-EAT, this research defines a taxonomy of EAT patterns describing the nine possible outcomes of an embedding association test, each of which is associated with a unique EAT-Map, a novel four-quadrant visualization for interpreting the ML-EAT. Empirical analysis of static and diachronic word embeddings, GPT-2 language models, and a CLIP language-and-image model shows that EAT patterns add otherwise unobservable information about the component biases that make up an EAT; reveal the effects of prompting in zero-shot models; and can also identify situations when cosine similarity is an ineffective metric, rendering an EAT unreliable. Our work contributes a method for rendering bias more observable and interpretable, improving the transparency of computational investigations into human minds and societies., Comment: Accepted at Artificial Intelligence, Ethics, and Society 2024

Popular and news media often portray teenagers with sensationalism, as both a risk to society and at risk from society. As AI begins to absorb some of the epistemic functions of traditional media, we study how teenagers in two countries speaking two languages: 1) are depicted by AI, and 2) how they would prefer to be depicted. Specifically, we study the biases about teenagers learned by static word embeddings (SWEs) and generative language models (GLMs), comparing these with the perspectives of adolescents living in the U.S. and Nepal. We find English-language SWEs associate teenagers with societal problems, and more than 50% of the 1,000 words most associated with teenagers in the pretrained GloVe SWE reflect such problems. Given prompts about teenagers, 30% of outputs from GPT2-XL and 29% from LLaMA-2-7B GLMs discuss societal problems, most commonly violence, but also drug use, mental illness, and sexual taboo. Nepali models, while not free of such associations, are less dominated by social problems. Data from workshops with N=13 U.S. adolescents and N=18 Nepalese adolescents show that AI presentations are disconnected from teenage life, which revolves around activities like school and friendship. Participant ratings of how well 20 trait words describe teens are decorrelated from SWE associations, with Pearson's r=.02, n.s. in English FastText and r=.06, n.s. in GloVe; and r=.06, n.s. in Nepali FastText and r=-.23, n.s. in GloVe. U.S. participants suggested AI could fairly present teens by highlighting diversity, while Nepalese participants centered positivity. Participants were optimistic that, if it learned from adolescents, rather than media sources, AI could help mitigate stereotypes. Our work offers an understanding of the ways SWEs and GLMs misrepresent a developmentally vulnerable group and provides a template for less sensationalized characterization., Comment: Accepted at Artificial Intelligence, Ethics, and Society 2024

Multimodal AI models capable of associating images and text hold promise for numerous domains, ranging from automated image captioning to accessibility applications for blind and low-vision users. However, uncertainty about bias has in some cases limited their adoption and availability. In the present work, we study 43 CLIP vision-language models to determine whether they learn human-like facial impression biases, and we find evidence that such biases are reflected across three distinct CLIP model families. We show for the first time that the the degree to which a bias is shared across a society predicts the degree to which it is reflected in a CLIP model. Human-like impressions of visually unobservable attributes, like trustworthiness and sexuality, emerge only in models trained on the largest dataset, indicating that a better fit to uncurated cultural data results in the reproduction of increasingly subtle social biases. Moreover, we use a hierarchical clustering approach to show that dataset size predicts the extent to which the underlying structure of facial impression bias resembles that of facial impression bias in humans. Finally, we show that Stable Diffusion models employing CLIP as a text encoder learn facial impression biases, and that these biases intersect with racial biases in Stable Diffusion XL-Turbo. While pretrained CLIP models may prove useful for scientific studies of bias, they will also require significant dataset curation when intended for use as general-purpose models in a zero-shot setting., Comment: Accepted at Artificial Intelligence, Ethics, and Society 2024

Equitable urban transportation applications require high-fidelity digital representations of the built environment: not just streets and sidewalks, but bike lanes, marked and unmarked crossings, curb ramps and cuts, obstructions, traffic signals, signage, street markings, potholes, and more. Direct inspections and manual annotations are prohibitively expensive at scale. Conventional machine learning methods require substantial annotated training data for adequate performance. In this paper, we consider vision language models as a mechanism for annotating diverse urban features from satellite images, reducing the dependence on human annotation to produce large training sets. While these models have achieved impressive results in describing common objects in images captured from a human perspective, their training sets are less likely to include strong signals for esoteric features in the built environment, and their performance in these settings is therefore unclear. We demonstrate proof-of-concept combining a state-of-the-art vision language model and variants of a prompting strategy that asks the model to consider segmented elements independently of the original image. Experiments on two urban features -- stop lines and raised tables -- show that while direct zero-shot prompting correctly annotates nearly zero images, the pre-segmentation strategies can annotate images with near 40% intersection-over-union accuracy. We describe how these results inform a new research agenda in automatic annotation of the built environment to improve equity, accessibility, and safety at broad scale and in diverse environments.

Abstention, the refusal of large language models (LLMs) to provide an answer, is increasingly recognized for its potential to mitigate hallucinations and enhance safety in LLM systems. In this survey, we introduce a framework to examine abstention from three perspectives: the query, the model, and human values. We organize the literature on abstention methods, benchmarks, and evaluation metrics using this framework, and discuss merits and limitations of prior work. We further identify and motivate areas for future work, centered around whether abstention can be achieved as a meta-capability that transcends specific tasks or domains, while still providing opportunities to optimize abstention abilities based on context., Comment: preprint

Applications to support pedestrian mobility in urban areas require a complete, and routable graph representation of the built environment. Globally available information, including aerial imagery provides a scalable source for constructing these path networks, but the associated learning problem is challenging: Relative to road network pathways, pedestrian network pathways are narrower, more frequently disconnected, often visually and materially variable in smaller areas, and their boundaries are broken up by driveway incursions, alleyways, marked or unmarked crossings through roadways. Existing algorithms to extract pedestrian pathway network graphs are inconsistently evaluated and tend to ignore routability, making it difficult to assess utility for mobility applications: Even if all path segments are available, discontinuities could dramatically and arbitrarily shift the overall path taken by a pedestrian. In this paper, we describe a first standard benchmark for the pedestrian pathway graph extraction problem, comprising the largest available dataset equipped with manually vetted ground truth annotations (covering $3,000 km^2$ land area in regions from 8 cities), and a family of evaluation metrics centering routability and downstream utility. By partitioning the data into polygons at the scale of individual intersections, we compute local routability as an efficient proxy for global routability. We consider multiple measures of polygon-level routability and compare predicted measures with ground truth to construct evaluation metrics. Using these metrics, we show that this benchmark can surface strengths and weaknesses of existing methods that are hidden by simple edge-counting metrics over single-region datasets used in prior work, representing a challenging, high-impact problem in computer vision and machine learning., Comment: arXiv admin note: text overlap with arXiv:2303.02323

The rapid proliferation of generative AI has raised questions about the competitiveness of lower-parameter, locally tunable, open-weight models relative to high-parameter, API-guarded, closed-weight models in terms of performance, domain adaptation, cost, and generalization. Centering under-resourced yet risk-intolerant settings in government, research, and healthcare, we see for-profit closed-weight models as incompatible with requirements for transparency, privacy, adaptability, and standards of evidence. Yet the performance penalty in using open-weight models, especially in low-data and low-resource settings, is unclear. We assess the feasibility of using smaller, open-weight models to replace GPT-4-Turbo in zero-shot, few-shot, and fine-tuned regimes, assuming access to only a single, low-cost GPU. We assess value-sensitive issues around bias, privacy, and abstention on three additional tasks relevant to those topics. We find that with relatively low effort, very low absolute monetary cost, and relatively little data for fine-tuning, small open-weight models can achieve competitive performance in domain-adapted tasks without sacrificing generality. We then run experiments considering practical issues in bias, privacy, and hallucination risk, finding that open models offer several benefits over closed models. We intend this work as a case study in understanding the opportunity cost of reproducibility and transparency over for-profit state-of-the-art zero shot performance, finding this cost to be marginal under realistic settings., Comment: Accepted at the ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2024

The correct model response in the face of uncertainty is to abstain from answering a question so as not to mislead the user. In this work, we study the ability of LLMs to abstain from answering context-dependent science questions when provided insufficient or incorrect context. We probe model sensitivity in several settings: removing gold context, replacing gold context with irrelevant context, and providing additional context beyond what is given. In experiments on four QA datasets with six LLMs, we show that performance varies greatly across models, across the type of context provided, and also by question type; in particular, many LLMs seem unable to abstain from answering boolean questions using standard QA prompts. Our analysis also highlights the unexpected impact of abstention performance on QA task accuracy. Counter-intuitively, in some settings, replacing gold context with irrelevant context or adding irrelevant context to gold context can improve abstention performance in a way that results in improvements in task performance. Our results imply that changes are needed in QA dataset design and evaluation to more effectively assess the correctness and downstream impacts of model abstention., Comment: EMNLP 2024 Findings

In this paper, we build a visual dialogue dataset, named InfoVisDial, which provides rich informative answers in each round even with external knowledge related to the visual content. Different from existing datasets where the answer is compact and short, InfoVisDial contains long free-form answers with rich information in each round of dialogue. For effective data collection, the key idea is to bridge the large-scale multimodal model (e.g., GIT) and the language models (e.g., GPT-3). GIT can describe the image content even with scene text, while GPT-3 can generate informative dialogue based on the image description and appropriate prompting techniques. With such automatic pipeline, we can readily generate informative visual dialogue data at scale. Then, we ask human annotators to rate the generated dialogues to filter the low-quality conversations.Human analyses show that InfoVisDial covers informative and diverse dialogue topics: $54.4\%$ of the dialogue rounds are related to image scene texts, and $36.7\%$ require external knowledge. Each round's answer is also long and open-ended: $87.3\%$ of answers are unique with an average length of $8.9$, compared with $27.37\%$ and $2.9$ in VisDial. Last, we propose a strong baseline by adapting the GIT model for the visual dialogue task and fine-tune the model on InfoVisDial. Hopefully, our work can motivate more effort on this direction.

An invasive species of grass known as "buffelgrass" contributes to severe wildfires and biodiversity loss in the Southwest United States. We tackle the problem of predicting buffelgrass "green-ups" (i.e. readiness for herbicidal treatment). To make our predictions, we explore temporal, visual and multi-modal models that combine satellite sensing and deep learning. We find that all of our neural-based approaches improve over conventional buffelgrass green-up models, and discuss how neural model deployment promises significant resource savings.

Open data is frequently released spatially aggregated, usually to comply with privacy policies. But coarse, heterogeneous aggregations complicate learning and integration for downstream AI/ML systems. In this work, we consider models to disaggregate spatio-temporal data from a low-resolution, irregular partition (e.g., census tract) to a high-resolution, irregular partition (e.g., city block). We propose an overarching model named the Structurally-Aware Recurrent Network (SARN), which integrates structurally-aware spatial attention (SASA) layers into the Gated Recurrent Unit (GRU) model. The spatial attention layers capture spatial interactions among regions, while the gated recurrent module captures the temporal dependencies. Each SASA layer calculates both global and structural attention -- global attention facilitates comprehensive interactions between different geographic levels, while structural attention leverages the containment relationship between different geographic levels (e.g., a city block being wholly contained within a census tract) to ensure coherent and consistent results. For scenarios with limited historical training data, we explore transfer learning and show that a model pre-trained on one city variable can be fine-tuned for another city variable using only a few hundred samples. Evaluating these techniques on two mobility datasets, we find that on both datasets, SARN significantly outperforms other neural models (5% and 1%) and typical heuristic methods (40% and 14%), enabling us to generate realistic, high-quality fine-grained data for downstream applications.

We adapt image inpainting techniques to impute large, irregular missing regions in urban settings characterized by sparsity, variance in both space and time, and anomalous events. Missing regions in urban data can be caused by sensor or software failures, data quality issues, interference from weather events, incomplete data collection, or varying data use regulations; any missing data can render the entire dataset unusable for downstream applications. To ensure coverage and utility, we adapt computer vision techniques for image inpainting to operate on 3D histograms (2D space + 1D time) commonly used for data exchange in urban settings. Adapting these techniques to the spatiotemporal setting requires handling skew: urban data tend to follow population density patterns (small dense regions surrounded by large sparse areas); these patterns can dominate the learning process and fool the model into ignoring local or transient effects. To combat skew, we 1) train simultaneously in space and time, and 2) focus attention on dense regions by biasing the masks used for training to the skew in the data. We evaluate the core model and these two extensions using the NYC taxi data and the NYC bikeshare data, simulating different conditions for missing data. We show that the core model is effective qualitatively and quantitatively, and that biased masking during training reduces error in a variety of scenarios. We also articulate a tradeoff in varying the number of timesteps per training sample: too few timesteps and the model ignores transient events; too many timesteps and the model is slow to train with limited performance gain.

Nine language-vision AI models trained on web scrapes with the Contrastive Language-Image Pretraining (CLIP) objective are evaluated for evidence of a bias studied by psychologists: the sexual objectification of girls and women, which occurs when a person's human characteristics, such as emotions, are disregarded and the person is treated as a body. We replicate three experiments in psychology quantifying sexual objectification and show that the phenomena persist in AI. A first experiment uses standardized images of women from the Sexual OBjectification and EMotion Database, and finds that human characteristics are disassociated from images of objectified women: the model's recognition of emotional state is mediated by whether the subject is fully or partially clothed. Embedding association tests (EATs) return significant effect sizes for both anger (d >0.80) and sadness (d >0.50), associating images of fully clothed subjects with emotions. GRAD-CAM saliency maps highlight that CLIP gets distracted from emotional expressions in objectified images. A second experiment measures the effect in a representative application: an automatic image captioner (Antarctic Captions) includes words denoting emotion less than 50% as often for images of partially clothed women than for images of fully clothed women. A third experiment finds that images of female professionals (scientists, doctors, executives) are likely to be associated with sexual descriptions relative to images of male professionals. A fourth experiment shows that a prompt of "a [age] year old girl" generates sexualized images (as determined by an NSFW classifier) up to 73% of the time for VQGAN-CLIP and Stable Diffusion; the corresponding rate for boys never surpasses 9%. The evidence indicates that language-vision AI models trained on web scrapes learn biases of sexual objectification, which propagate to downstream applications., Comment: 12 pages, 4 figures, 2 tables

Differential privacy (DP) data synthesizers support public release of sensitive information, offering theoretical guarantees for privacy but limited evidence of utility in practical settings. Utility is typically measured as the error on representative proxy tasks, such as descriptive statistics, accuracy of trained classifiers, or performance over a query workload. The ability for these results to generalize to practitioners' experience has been questioned in a number of settings, including the U.S. Census. In this paper, we propose an evaluation methodology for synthetic data that avoids assumptions about the representativeness of proxy tasks, instead measuring the likelihood that published conclusions would change had the authors used synthetic data, a condition we call epistemic parity. Our methodology consists of reproducing empirical conclusions of peer-reviewed papers on real, publicly available data, then re-running these experiments a second time on DP synthetic data, and comparing the results. We instantiate our methodology over a benchmark of recent peer-reviewed papers that analyze public datasets in the ICPSR repository. We model quantitative claims computationally to automate the experimental workflow, and model qualitative claims by reproducing visualizations and comparing the results manually. We then generate DP synthetic datasets using multiple state-of-the-art mechanisms, and estimate the likelihood that these conclusions will hold. We find that state-of-the-art DP synthesizers are able to achieve high epistemic parity for several papers in our benchmark. However, some papers, and particularly some specific findings, are difficult to reproduce for any of the synthesizers. We advocate for a new class of mechanisms that favor stronger utility guarantees and offer privacy protection with a focus on application-specific threat models and risk-assessment., Comment: Preprint. 14 pages

Machine learning algorithms are commonly specified in linear algebra (LA). LA expressions can be rewritten into more efficient forms, by taking advantage of input properties such as sparsity, as well as program properties such as common subexpressions and fusible operators. The complex interaction among these properties' impact on the execution cost poses a challenge to optimizing compilers. Existing compilers resort to intricate heuristics that complicate the codebase and add maintenance cost but fail to search through the large space of equivalent LA expressions to find the cheapest one. We introduce a general optimization technique for LA expressions, by converting the LA expressions into Relational Algebra (RA) expressions, optimizing the latter, then converting the result back to (optimized) LA expressions. One major advantage of this method is that it is complete, meaning that any equivalent LA expression can be found using the equivalence rules in RA. The challenge is the major size of the search space, and we address this by adopting and extending a technique used in compilers, called equality saturation. We integrate the optimizer into SystemML and validate it empirically across a spectrum of machine learning tasks; we show that we can derive all existing hand-coded optimizations in SystemML, and perform new optimizations that lead to speedups from 1.2X to 5X.

Fairness is increasingly recognized as a critical component of machine learning systems. However, it is the underlying data on which these systems are trained that often reflects discrimination, suggesting a data management problem. In this paper, we first make a distinction between associational and causal definitions of fairness in the literature and argue that the concept of fairness requires causal reasoning. We then review existing works and identify future opportunities for applying data management techniques to causal algorithmic fairness., Comment: arXiv admin note: text overlap with arXiv:1902.08283

Figures are an important channel for scientific communication, used to express complex ideas, models and data in ways that words cannot. However, this visual information is mostly ignored in analyses of the scientific literature. In this paper, we demonstrate the utility of using scientific figures as markers of knowledge domains in science, which can be used for classification, recommender systems, and studies of scientific information exchange. We encode sets of images into a visual signature, then use distances between these signatures to understand how patterns of visual communication compare with patterns of jargon and citation structures. We find that figures can be as effective for differentiating communities of practice as text or citation patterns. We then consider where these metrics disagree to understand how different disciplines use visualization to express ideas. Finally, we further consider how specific figure types propagate through the literature, suggesting a new mechanism for understanding the flow of ideas apart from conventional channels of text and citations. Our ultimate aim is to better leverage these information-dense objects to improve scientific communication across disciplinary boundaries.

Emerging transportation modes, including car-sharing, bike-sharing, and ride-hailing, are transforming urban mobility but have been shown to reinforce socioeconomic inequities. Spatiotemporal demand prediction models for these new mobility regimes must therefore consider fairness as a first-class design requirement. We present FairST, a fairness-aware model for predicting demand for new mobility systems. Our approach utilizes 1D, 2D and 3D convolutions to integrate various urban features and learn the spatial-temporal dynamics of a mobility system, but we include fairness metrics as a form of regularization to make the predictions more equitable across demographic groups. We propose two novel spatiotemporal fairness metrics, a region-based fairness gap (RFG) and an individual-based fairness gap (IFG). Both quantify equity in a spatiotemporal context, but vary by whether demographics are labeled at the region level (RFG) or whether population distribution information is available (IFG). Experimental results on real bike share and ride share datasets demonstrate the effectiveness of the proposed model: FairST not only reduces the fairness gap by more than 80%, but can surprisingly achieve better accuracy than state-of-the-art yet fairness-oblivious methods including LSTMs, ConvLSTMs, and 3D CNN.

Synthetic datasets have long been thought of as second-rate, to be used only when "real" data collected directly from the real world is unavailable. But this perspective assumes that raw data is clean, unbiased, and trustworthy, which it rarely is. Moreover, the benefits of synthetic data for privacy and for bias correction are becoming increasingly important in any domain that works with people. Curated synthetic datasets - synthetic data derived from minimal perturbations of real data - enable early stage product development and collaboration, protect privacy, afford reproducibility, increase dataset diversity in research, and protect disadvantaged groups from problematic inferences on the original data that reflects systematic discrimination. Rather than representing a departure from the true state of the world, in this paper we argue that properly generated synthetic data is a step towards responsible and equitable research and development of machine learning systems., Comment: Discussion paper at FATES on the Web 2019

Fairness is increasingly recognized as a critical component of machine learning systems. However, it is the underlying data on which these systems are trained that often reflect discrimination, suggesting a database repair problem. Existing treatments of fairness rely on statistical correlations that can be fooled by statistical anomalies, such as Simpson's paradox. Proposals for causality-based definitions of fairness can correctly model some of these situations, but they require specification of the underlying causal models. In this paper, we formalize the situation as a database repair problem, proving sufficient conditions for fair classifiers in terms of admissible variables as opposed to a complete causal model. We show that these conditions correctly capture subtle fairness violations. We then use these conditions as the basis for database repair algorithms that provide provable fairness guarantees about classifiers trained on their training labels. We evaluate our algorithms on real data, demonstrating improvement over the state of the art on multiple fairness metrics proposed in the literature while retaining high utility.

We propose JECL, a method for clustering image-caption pairs by training parallel encoders with regularized clustering and alignment objectives, simultaneously learning both representations and cluster assignments. These image-caption pairs arise frequently in high-value applications where structured training data is expensive to produce, but free-text descriptions are common. JECL trains by minimizing the Kullback-Leibler divergence between the distribution of the images and text to that of a combined joint target distribution and optimizing the Jensen-Shannon divergence between the soft cluster assignments of the images and text. Regularizers are also applied to JECL to prevent trivial solutions. Experiments show that JECL outperforms both single-view and multi-view methods on large benchmark image-caption datasets, and is remarkably robust to missing captions and varying data sizes., Comment: ICPR2020

We present a system to support generalized SQL workload analysis and management for multi-tenant and multi-database platforms. Workload analysis applications are becoming more sophisticated to support database administration, model user behavior, audit security, and route queries, but the methods rely on specialized feature engineering, and therefore must be carefully implemented and reimplemented for each SQL dialect, database system, and application. Meanwhile, the size and complexity of workloads are increasing as systems centralize in the cloud. We model workload analysis and management tasks as variations on query labeling, and propose a system design that can support general query labeling routines across multiple applications and database backends. The design relies on the use of learned vector embeddings for SQL queries as a replacement for application-specific syntactic features, reducing custom code and allowing the use of off-the-shelf machine learning algorithms for labeling. The key hypothesis, for which we provide evidence in this paper, is that these learned features can outperform conventional feature engineering on representative machine learning tasks. We present the design of a database-agnostic workload management and analytics service, describe potential applications, and show that separating workload representation from labeling tasks affords new capabilities and can outperform existing solutions for representative tasks, including workload sampling for index recommendation and user labeling for security audits.

Techniques to deliver privacy-preserving synthetic datasets take a sensitive dataset as input and produce a similar dataset as output while maintaining differential privacy. These approaches have the potential to improve data sharing and reuse, but they must be accessible to non-experts and tolerant of realistic data. Existing approaches make an implicit assumption that the active domain of the dataset is similar to the global domain, potentially violating differential privacy. In this paper, we present an algorithm for generating differentially private synthetic data over the large, weakly constrained domains we find in realistic open data situations. Our algorithm models the unrepresented domain analytically as a probability distribution to adjust the output and compute noise, avoiding the need to compute the full domain explicitly. We formulate the tradeoff between privacy and utility in terms of a "tolerance for randomness" parameter that does not require users to inspect the data to set. Finally, we show that the algorithm produces sensible results on real datasets., Comment: Submitted to TDPD18

We describe customized synthetic datasets for publishing mobility data. Private companies are providing new transportation modalities, and their data is of high value for integrative transportation research, policy enforcement, and public accountability. However, these companies are disincentivized from sharing data not only to protect the privacy of individuals (drivers and/or passengers), but also to protect their own competitive advantage. Moreover, demographic biases arising from how the services are delivered may be amplified if released data is used in other contexts. We describe a model and algorithm for releasing origin-destination histograms that removes selected biases in the data using causality-based methods. We compute the origin-destination histogram of the original dataset then adjust the counts to remove undesirable causal relationships that can lead to discrimination or violate contractual obligations with data owners. We evaluate the utility of the algorithm on real data from a dockless bike share program in Seattle and taxi data in New York, and show that these adjusted transportation datasets can retain utility while removing bias in the underlying data., Comment: Presented at BIDU 2018 workshop and published in Springer Communications in Computer and Information Science vol 926

Algorithmic decisions often result in scoring and ranking individuals to determine credit worthiness, qualifications for college admissions and employment, and compatibility as dating partners. While automatic and seemingly objective, ranking algorithms can discriminate against individuals and protected groups, and exhibit low diversity. Furthermore, ranked results are often unstable --- small changes in the input data or in the ranking methodology may lead to drastic changes in the output, making the result uninformative and easy to manipulate. Similar concerns apply in cases where items other than individuals are ranked, including colleges, academic departments, or products. In this demonstration we present Ranking Facts, a Web-based application that generates a "nutritional label" for rankings. Ranking Facts is made up of a collection of visual widgets that implement our latest research results on fairness, stability, and transparency for rankings, and that communicate details of the ranking methodology, or of the output, to the end user. We will showcase Ranking Facts on real datasets from different domains, including college rankings, criminal risk assessment, and financial services., Comment: 4 pages, SIGMOD demo, 3 figuress, ACM SIGMOD 2018

We consider methods for learning vector representations of SQL queries to support generalized workload analytics tasks, including workload summarization for index selection and predicting queries that will trigger memory errors. We consider vector representations of both raw SQL text and optimized query plans, and evaluate these methods on synthetic and real SQL workloads. We find that general algorithms based on vector representations can outperform existing approaches that rely on specialized features. For index recommendation, we cluster the vector representations to compress large workloads with no loss in performance from the recommended index. For error prediction, we train a classifier over learned vectors that can automatically relate subtle syntactic patterns with specific errors raised during query execution. Surprisingly, we also find that these methods enable transfer learning, where a model trained on one SQL corpus can be applied to an unrelated corpus and still enable good performance. We find that these general approaches, when trained on a large corpus of SQL queries, provides a robust foundation for a variety of workload analysis tasks and database features, without requiring application-specific feature engineering.

Data for good implies unfettered access to data. But data owners must be conservative about how, when, and why they share data or risk violating the trust of the people they aim to help, losing their funding, or breaking the law. Data sharing agreements can help prevent privacy violations, but require a level of specificity that is premature during preliminary discussions, and can take over a year to establish. We consider the generation and use of synthetic data to facilitate ad hoc collaborations involving sensitive data. A good synthetic dataset has two properties: it is representative of the original data, and it provides strong guarantees about privacy. In this paper, we discuss important use cases for synthetic data that challenge the state of the art in privacy-preserving data generation, and describe DataSynthesizer, a dataset generation tool that takes a sensitive dataset as input and generates a structurally and statistically similar synthetic dataset, with strong privacy guarantees, as output. The data owners need not release their data, while potential collaborators can begin developing models and methods with some confidence that their results will work similarly on the real dataset. The distinguishing feature of DataSynthesizer is its usability - in most cases, the data owner need not specify any parameters to start generating and sharing data safely and effectively. The code implementing DataSynthesizer is publicly available on GitHub at https://github.com/DataResponsibly. The work on DataSynthesizer is part of the Data, Responsibly project, where the goal is to operationalize responsibility in data sharing, integration, analysis and use., Comment: Presented at the Data For Good Exchange 2017

Many real-world applications require automated data annotation, such as identifying tissue origins based on gene expressions and classifying images into semantic categories. Annotation classes are often numerous and subject to changes over time, and annotating examples has become the major bottleneck for supervised learning methods. In science and other high-value domains, large repositories of data samples are often available, together with two sources of organic supervision: a lexicon for the annotation classes, and text descriptions that accompany some data samples. Distant supervision has emerged as a promising paradigm for exploiting such indirect supervision by automatically annotating examples where the text description contains a class mention in the lexicon. However, due to linguistic variations and ambiguities, such training data is inherently noisy, which limits the accuracy of this approach. In this paper, we introduce an auxiliary natural language processing system for the text modality, and incorporate co-training to reduce noise and augment signal in distant supervision. Without using any manually labeled data, our EZLearn system learned to accurately annotate data samples in functional genomics and scientific figure comprehension, substantially outperforming state-of-the-art supervised methods trained on tens of thousands of annotated examples.

Analytics tasks manipulate structured data with variants of relational algebra (RA) and quantitative data with variants of linear algebra (LA). The two computational models have overlapping expressiveness, motivating a common programming model that affords unified reasoning and algorithm design. At the logical level we propose Lara, a lean algebra of three operators, that expresses RA and LA as well as relevant optimization rules. We show a series of proofs that position Lara %formal and informal at just the right level of expressiveness for a middleware algebra: more explicit than MapReduce but more general than RA or LA. At the physical level we find that the Lara operators afford efficient implementations using a single primitive that is available in a variety of backend engines: range scans over partitioned sorted maps. To evaluate these ideas, we implemented the Lara operators as range iterators in Apache Accumulo, a popular implementation of Google's BigTable. First we show how Lara expresses a sensor quality control task, and we measure the performance impact of optimizations Lara admits on this task. Second we show that the LaraDB implementation outperforms Accumulo's native MapReduce integration on a core task involving join and aggregation in the form of matrix multiply, especially at smaller scales that are typically a poor fit for scale-out approaches. We find that LaraDB offers a conceptually lean framework for optimizing mixed-abstraction analytics tasks, without giving up fast record-level updates and scans., Comment: 10 pages, to appear in the BeyondMR workshop at the 2017 ACM SIGMOD conference

Google BigTable's scale-out design for distributed key-value storage inspired a generation of NoSQL databases. Recently the NewSQL paradigm emerged in response to analytic workloads that demand distributed computation local to data storage. Many such analytics take the form of graph algorithms, a trend that motivated the GraphBLAS initiative to standardize a set of matrix math kernels for building graph algorithms. In this article we show how it is possible to implement the GraphBLAS kernels in a BigTable database by presenting the design of Graphulo, a library for executing graph algorithms inside the Apache Accumulo database. We detail the Graphulo implementation of two graph algorithms and conduct experiments comparing their performance to two main-memory matrix math systems. Our results shed insight into the conditions that determine when executing a graph algorithm is faster inside a database versus an external system---in short, that memory requirements and relative I/O are critical factors., Comment: 9 pages, to appear in 2016 IEEE High Performance Extreme Computing Conference (HPEC)

Scientific results are communicated visually in the literature through diagrams, visualizations, and photographs. These information-dense objects have been largely ignored in bibliometrics and scientometrics studies when compared to citations and text. In this paper, we use techniques from computer vision and machine learning to classify more than 8 million figures from PubMed into 5 figure types and study the resulting patterns of visual information as they relate to impact. We find that the distribution of figures and figure types in the literature has remained relatively constant over time, but can vary widely across field and topic. Remarkably, we find a significant correlation between scientific impact and the use of visual information, where higher impact papers tend to include more diagrams, and to a lesser extent more plots and photographs. To explore these results and other ways of extracting this visual information, we have built a visual browser to illustrate the concept and explore design alternatives for supporting viziometric analysis and organizing visual information. We use these results to articulate a new research agenda -- viziometrics -- to study the organization and presentation of visual information in the scientific literature.

Data processing systems roughly group into families such as relational, array, graph, and key-value. Many data processing tasks exceed the capabilities of any one family, require data stored across families, or run faster when partitioned onto multiple families. Discovering ways to execute computation among multiple available systems, let alone discovering an optimal execution plan, is challenging given semantic differences between disparate families of systems. In this paper we introduce a new algebra, Lara, which underlies and unifies algebras representing the families above in order to facilitate translation between systems. We describe the operations and objects of Lara---union, join, and ext on associative tables---and show her properties and equivalences to other algebras. Multi-system optimization has a bright future, in which we proffer Lara for the role of universal connector., Comment: Working draft

In the coming decade, astronomical surveys of the sky will generate tens of terabytes of images and detect hundreds of millions of sources every night. The study of these sources will involve computation challenges such as anomaly detection and classification, and moving object tracking. Since such studies benefit from the highest quality data, methods such as image coaddition (stacking) will be a critical preprocessing step prior to scientific investigation. With a requirement that these images be analyzed on a nightly basis to identify moving sources or transient objects, these data streams present many computational challenges. Given the quantity of data involved, the computational load of these problems can only be addressed by distributing the workload over a large number of nodes. However, the high data throughput demanded by these applications may present scalability challenges for certain storage architectures. One scalable data-processing method that has emerged in recent years is MapReduce, and in this paper we focus on its popular open-source implementation called Hadoop. In the Hadoop framework, the data is partitioned among storage attached directly to worker nodes, and the processing workload is scheduled in parallel on the nodes that contain the required input data. A further motivation for using Hadoop is that it allows us to exploit cloud computing resources, e.g., Amazon's EC2. We report on our experience implementing a scalable image-processing pipeline for the SDSS imaging database using Hadoop. This multi-terabyte imaging dataset provides a good testbed for algorithm development since its scope and structure approximate future surveys. First, we describe MapReduce and how we adapted image coaddition to the MapReduce framework. Then we describe a number of optimizations to our basic approach and report experimental results comparing their performance., Comment: 31 pages, 11 figures, 2 tables