Graph neural networks (GNNs) are widely used in urban spatiotemporal forecasting, e.g., predicting infrastructure problems. In this setting, government officials aim to identify in which neighborhoods incidents like potholes or rodents occur. The true state of incidents is observed via government inspection ratings. However, these ratings are only conducted for a sparse set of neighborhoods and incident types. We also observe the state of incidents via crowdsourced reports, which are more densely observed but may be biased due to heterogeneous reporting. First, we propose a multiview, multioutput GNN-based model that uses both unbiased rating data and biased reporting data to predict the true latent state of incidents. Second, we investigate a case study of New York City urban incidents and collect a dataset of 9,615,863 crowdsourced reports and 1,041,415 government inspection ratings over 3 years and across 139 types of incidents. We show on both real and semi-synthetic data that our model can better predict the latent state compared to models that use only reporting data or only rating data. Finally, we quantify demographic biases in crowdsourced reporting, e.g., higher-income neighborhoods report problems at higher rates. Our analysis showcases a widely applicable approach for latent state prediction using heterogeneous, sparse, and biased data.

We describe HypotheSAEs, a general method to hypothesize interpretable relationships between text data (e.g., headlines) and a target variable (e.g., clicks). HypotheSAEs has three steps: (1) train a sparse autoencoder on text embeddings to produce interpretable features describing the data distribution, (2) select features that predict the target variable, and (3) generate a natural language interpretation of each feature (e.g., mentions being surprised or shocked) using an LLM. Each interpretation serves as a hypothesis about what predicts the target variable. Compared to baselines, our method better identifies reference hypotheses on synthetic datasets (at least +0.06 in F1) and produces more predictive hypotheses on real datasets ( twice as many significant findings), despite requiring 1-2 orders of magnitude less compute than recent LLM-based methods. HypotheSAEs also produces novel discoveries on two well-studied tasks: explaining partisan differences in Congressional speeches and identifying drivers of engagement with online headlines.

Generating social networks is essential for many applications, such as epidemic modeling and social simulations. The emergence of generative AI, especially large language models (LLMs), offers new possibilities for social network generation: LLMs can generate networks without additional training or need to define network parameters, and users can flexibly define individuals in the network using natural language. However, this potential raises two critical questions: 1) are the social networks generated by LLMs realistic, and 2) what are risks of bias, given the importance of demographics in forming social ties? To answer these questions, we develop three prompting methods for network generation and compare the generated networks to a suite of real social networks. We find that more realistic networks are generated with “local” methods, where the LLM constructs relations for one persona at a time, compared to “global” methods that construct the entire network at once. We also find that the generated networks match real networks on many characteristics, including density, clustering, connectivity, and degree distribution. However, we find that LLMs emphasize political homophily over all other types of homophily and significantly overestimate political homophily compared to real social networks.

It is difficult to evaluate machine learning classifiers without large labeled datasets, which are often unavailable. In contrast, unlabeled data is plentiful, but not easily used for evaluation. Here, we introduce Semi-Supervised Model Evaluation (SSME), a method that uses both labeled and unlabeled data to evaluate machine learning classifiers. The key idea is to estimate the joint distribution of ground truth labels and classifier scores using a semi-supervised mixture model. The semi-supervised mixture model allows SSME to learn from three sources of information: unlabeled data, multiple classifiers, and probabilistic classifier scores. Once fit, the mixture model enables estimation of any metric that is a function of classifier scores and ground truth labels (e.g., accuracy or AUC). We derive theoretical bounds on the error of these estimates, showing that estimation error decreases with the number of classifiers and the amount of unlabeled data. We present experiments in four domains where obtaining large labeled datasets is often impractical: healthcare, content moderation, molecular property prediction, and text classification. Our results demonstrate that SSME estimates performance more accurately than do competing methods, reducing error by 5.1x relative to using labeled data alone and 2.4x relative to the next best method.

Decision-makers often observe the occurrence of events through a reporting process. City governments, for example, rely on resident reports to find and then resolve urban infrastructural problems such as fallen street trees, flooded basements, or rat infestations. Without additional assumptions, there is no way to distinguish events that occur but are not reported from events that truly did not occur--a fundamental problem in settings with positive-unlabeled data. Because disparities in reporting rates correlate with resident demographics, addressing incidents only on the basis of reports leads to systematic neglect in neighborhoods that are less likely to report events. We show how to overcome this challenge by leveraging the fact that events are spatially correlated. Our framework uses a Bayesian spatial latent variable model to infer event occurrence probabilities and applies it to storm-induced flooding reports in New York City, further pooling results across multiple storms. We show that a model accounting for under-reporting and spatial correlation predicts future reports more accurately than other models, and further induces a more equitable set of inspections: its allocations better reflect the population and provide equitable service to non-white, less traditionally educated, and lower-income residents. This finding reflects heterogeneous reporting behavior learned by the model: reporting rates are higher in Census tracts with higher populations, proportions of white residents, and proportions of owner-occupied households. Our work lays the groundwork for more equitable proactive government services, even with disparate reporting behavior.

Do LLMs align with human perceptions of safety?We study this question via annotation alignment, the extent to which LLMs and humans agree when annotating the safety of userchatbot conversations.We leverage the recent DICES dataset (Aroyo et al., 2023), in which 350 conversations are each rated for safety by 112 annotators spanning 10 race-gender groups.GPT-4 achieves a Pearson correlation of r = 0.59 with the average annotator rating, higher than the median annotator's correlation with the average (r = 0.51).We show that larger datasets are needed to resolve whether GPT-4 exhibits disparities in how well it correlates with different demographic groups.Also, there is substantial idiosyncratic variation in correlation within groups, suggesting that race & gender do not fully capture differences in alignment.Finally, we find that GPT-4 cannot predict when one demographic group finds a conversation more unsafe than another.

Machine learning models are often trained to predict the outcome resulting from a human decision. For example, if a doctor decides to test a patient for disease, will the patient test positive? A challenge is that historical decision-making determines whether the outcome is observed: we only observe test outcomes for patients doctors historically tested. Untested patients, for whom outcomes are unobserved, may differ from tested patients along observed and unobserved dimensions. We propose a Bayesian model class which captures this setting. The purpose of the model is to accurately estimate risk for both tested and untested patients. Estimating this model is challenging due to the wide range of possibilities for untested patients. To address this, we propose two domain constraints which are plausible in health settings: a prevalence constraint, where the overall disease prevalence is known, and an expertise constraint, where the human decision-maker deviates from purely risk-based decision-making only along a constrained feature set. We show theoretically and on synthetic data that domain constraints improve parameter inference. We apply our model to a case study of cancer risk prediction, showing that the model's inferred risk predicts cancer diagnoses, its inferred testing policy captures known public health policies, and it can identify suboptimalities in test allocation. Though our case study is in healthcare, our analysis reveals a general class of domain constraints which can improve model estimation in many settings.

Rajiv Movva, Sidhika Balachandar, Kenny Peng, Gabriel Agostini, Nikhil Garg, Emma Pierson. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Users typically engage with LLMs interactively, yet most existing benchmarks evaluate them in a static, single-turn format, posing reliability concerns in interactive scenarios. We identify a key obstacle towards reliability: LLMs are trained to answer any question, even with incomplete context or insufficient knowledge. In this paper, we propose to change the static paradigm to an interactive one, develop systems that proactively ask questions to gather more information and respond reliably, and introduce an benchmark—MEDIQ—to evaluate question-asking ability in LLMs. MEDIQ simulates clinical interactions consisting of a Patient System and an adaptive Expert System; with potentially incomplete initial information, the Expert refrains from making diagnostic decisions when unconfident, and instead elicits missing details via follow-up questions. We provide a pipeline to convert single-turn medical benchmarks into an interactive format. Our results show that directly prompting state-of-the-art LLMs to ask questions degrades performance, indicating that adapting LLMs to proactive information-seeking settings is nontrivial. We experiment with abstention strategies to better estimate model confidence and decide when to ask questions, improving diagnostic accuracy by 22.3%; however, performance still lags compared to an (unrealistic in practice) upper bound with complete information upfront. Further analyses show improved interactive performance with filtering irrelevant contexts and reformatting conversations. Overall, we introduce a novel problem towards LLM reliability, an interactive MEDIQ benchmark and a novel question-asking system, and highlight directions to extend LLMs’ information-seeking abilities in critical domains.

When making recommendations, there is an apparent trade-off between the goals of accuracy (to recommend items a user is most likely to want) and diversity (to recommend items representing a range of categories). As such, real-world recommender systems often explicitly incorporate diversity into recommendations, at the cost of accuracy.

Distribution shifts—where the training distribution differs from the test distribution—can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity in the real-world deployments, these distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated benchmark of 10 datasets reflecting a diverse range of distribution shifts that naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training yields substantially lower out-of-distribution than in-distribution performance. This gap remains even with models trained by existing methods for tackling distribution shifts, underscoring the need for new methods for training models that are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. The full paper, code, and leaderboards are available at https://wilds.stanford.edu.

Mobility restrictions have been a primary intervention for controlling the spread of COVID-19, but they also place a significant economic burden on individuals and businesses. To balance these competing demands, policymakers need analytical tools to assess the costs and benefits of different mobility reduction measures. In this paper, we present our work motivated by our interactions with the Virginia Department of Health on a decision-support tool that utilizes large-scale data and epidemiological modeling to quantify the impact of changes in mobility on infection rates. Our model captures the spread of COVID-19 by using a fine-grained, dynamic mobility network that encodes the hourly movements of people from neighborhoods to individual places, with over 3 billion hourly edges. By perturbing the mobility network, we can simulate a wide variety of reopening plans and forecast their impact in terms of new infections and the loss in visits per sector. To deploy this model in practice, we built a robust computational infrastructure to support running millions of model realizations, and we worked with policymakers to develop an interactive dashboard that communicates our model's predictions for thousands of potential policies.

We seek to learn models that we can interact with using high-level concepts: if the model did not think there was a bone spur in the x-ray, would it still predict severe arthritis? State-of-the-art models today do not typically support the manipulation of concepts like "the existence of bone spurs", as they are trained end-to-end to go directly from raw input (e.g., pixels) to output (e.g., arthritis severity). We revisit the classic idea of first predicting concepts that are provided at training time, and then using these concepts to predict the label. By construction, we can intervene on these concept bottleneck models by editing their predicted concept values and propagating these changes to the final prediction. On x-ray grading and bird identification, concept bottleneck models achieve competitive accuracy with standard end-to-end models, while enabling interpretation in terms of high-level clinical concepts ("bone spurs") or bird attributes ("wing color"). These models also allow for richer human-model interaction: accuracy improves significantly if we can correct model mistakes on concepts at test time.

Modeling how individuals evolve over time is a fundamental problem in the natural and social sciences. However, existing datasets are often cross-sectional with each individual observed only once, making it impossible to apply traditional time-series methods. Motivated by the study of human aging, we present an interpretable latent-variable model that learns temporal dynamics from cross-sectional data. Our model represents each individual’s features over time as a nonlinear function of a low-dimensional, linearly-evolving latent state. We prove that when this nonlinear function is constrained to be order-isomorphic, the model family is identifiable solely from cross-sectional data provided the distribution of time-independent variation is known. On the UK Biobank human health dataset, our model reconstructs the observed data while learning interpretable rates of aging associated with diseases, mortality, and aging risk factors.

Predicting pregnancy has been a fundamental problem in women's health for more than 50 years. Previous datasets have been collected via carefully curated medical studies, but the recent growth of women's health tracking mobile apps offers potential for reaching a much broader population. However, the feasibility of predicting pregnancy from mobile health tracking data is unclear. Here we develop four models - a logistic regression model, and 3 LSTM models - to predict a woman's probability of becoming pregnant using data from a women's health tracking app, Clue by BioWink GmbH. Evaluating our models on a dataset of 79 million logs from 65,276 women with ground truth pregnancy test data, we show that our predicted pregnancy probabilities meaningfully stratify women: women in the top 10% of predicted probabilities have a 89% chance of becoming pregnant over 6 menstrual cycles, as compared to a 27% chance for women in the bottom 10%. We develop a technique for extracting interpretable time trends from our deep learning models, and show these trends are consistent with previous fertility research. Our findings illustrate the potential that women's health tracking data offers for predicting pregnancy on a broader population; we conclude by discussing the steps needed to fulfill this potential.

Threshold tests have recently been proposed as a useful method for detecting bias in lending, hiring, and policing decisions. For example, in the case of credit extensions, these tests aim to estimate the bar for granting loans to white and minority applicants, with a higher inferred threshold for minorities indicative of discrimination. This technique, however, requires fitting a complex Bayesian latent variable model for which inference is often computationally challenging. Here we develop a method for fitting threshold tests that is two orders of magnitude faster than the existing approach, reducing computation from hours to minutes. To achieve these performance gains, we introduce and analyze a flexible family of probability distributions on the interval [0, 1] – which we call discriminant distributions – that is computationally efficient to work with. We demonstrate our technique by analyzing 2.7 million police stops of pedestrians in New York City.

Cycles are fundamental to human health and behavior. Examples include mood cycles, circadian rhythms, and the menstrual cycle. However, modeling cycles in time series data is challenging because in most cases the cycles are not labeled or directly observed and need to be inferred from multidimensional measurements taken over time. Here, we present Cyclic Hidden Markov Models (CyHMMs) for detecting and modeling cycles in a collection of multidimensional heterogeneous time series data. In contrast to previous cycle modeling methods, CyHMMs deal with a number of challenges encountered in modeling real-world cycles: they can model multivariate data with both discrete and continuous dimensions; they explicitly model and are robust to missing data; and they can share information across individuals to accommodate variation both within and between individual time series. Experiments on synthetic and real-world health-tracking data demonstrate that CyHMMs infer cycle lengths more accurately than existing methods, with 58% lower error on simulated data and 63% lower error on real-world data compared to the best-performing baseline. CyHMMs can also perform functions which baselines cannot: they can model the progression of individual features/symptoms over the course of the cycle, identify the most variable features, and cluster individual time series into groups with distinct characteristics. Applying CyHMMs to two real-world health-tracking datasets -- of human menstrual cycle symptoms and physical activity tracking data -- yields important insights including which symptoms to expect at each point during the cycle. We also find that people fall into several groups with distinct cycle patterns, and that these groups differ along dimensions not provided to the model. For example, by modeling missing data in the menstrual cycles dataset, we are able to discover a medically relevant group of birth control users even though information on birth control is not given to the model.

Algorithms are now regularly used to decide whether defendants awaiting trial are too dangerous to be released back into the community. In some cases, black defendants are substantially more likely than white defendants to be incorrectly classified as high risk. To mitigate such disparities, several techniques have recently been proposed to achieve algorithmic fairness. Here we reformulate algorithmic fairness as constrained optimization: the objective is to maximize public safety while satisfying formal fairness constraints designed to reduce racial disparities. We show that for several past definitions of fairness, the optimal algorithms that result require detaining defendants above race-specific risk thresholds. We further show that the optimal unconstrained algorithm requires applying a single, uniform threshold to all defendants. The unconstrained algorithm thus maximizes public safety while also satisfying one important understanding of equality: that all individuals are held to the same standard, irrespective of race. Because the optimal constrained and unconstrained algorithms generally differ, there is tension between improving public safety and satisfying prevailing notions of algorithmic fairness. By examining data from Broward County, Florida, we show that this trade-off can be large in practice. We focus on algorithms for pretrial release decisions, but the principles we discuss apply to other domains, and also to human decision makers carrying out structured decision rules.

MOTIVATION: Chromatin immune-precipitation sequencing (ChIP-seq) experiments are commonly used to obtain genome-wide profiles of histone modifications associated with different types of functional genomic elements. However, the quality of histone ChIP-seq data is affected by many experimental parameters such as the amount of input DNA, antibody specificity, ChIP enrichment and sequencing depth. Making accurate inferences from chromatin profiling experiments that involve diverse experimental parameters is challenging. RESULTS: We introduce a convolutional denoising algorithm, Coda, that uses convolutional neural networks to learn a mapping from suboptimal to high-quality histone ChIP-seq data. This overcomes various sources of noise and variability, substantially enhancing and recovering signal when applied to low-quality chromatin profiling datasets across individuals, cell types and species. Our method has the potential to improve data quality at reduced costs. More broadly, this approach-using a high-dimensional discriminative model to encode a generative noise process-is generally applicable to other biological domains where it is easy to generate noisy data but difficult to analytically characterize the noise or underlying data distribution. AVAILABILITY AND IMPLEMENTATION: https://github.com/kundajelab/coda . CONTACT: [email protected].

Using eight months of online comments on New York Times articles, we find that only 28% of commenters of identifiable gender are female, but that their comments receive more recommendations from other readers. Comments from women are more common on forums about parenting, fashion, and health, and on articles written by women. The number of recommendations comments from women receive is positively correlated with the percentage of men on a forum, and the number of recommendations men receive is negatively correlated with the percentage of men on a forum. Female commenters are more likely to remain anonymous and anonymous commenters receive fewer recommendations. Male and female commenters differ in their choice of topics to emphasize, backgrounds, and language; we find three specific examples in responses to articles about sexual assault, contraception, and farm subsidies. We discuss the implications of these gender differences for democratic discourse and suggest ways to increase gender parity.

To understand the regulation of tissue-specific gene expression, the GTEx Consortium generated RNA-seq expression data for more than thirty distinct human tissues. This data provides an opportunity for deriving shared and tissue specific gene regulatory networks on the basis of co-expression between genes. However, a small number of samples are available for a majority of the tissues, and therefore statistical inference of networks in this setting is highly underpowered. To address this problem, we infer tissue-specific gene co-expression networks for 35 tissues in the GTEx dataset using a novel algorithm, GNAT, that uses a hierarchy of tissues to share data between related tissues. We show that this transfer learning approach increases the accuracy with which networks are learned. Analysis of these networks reveals that tissue-specific transcription factors are hubs that preferentially connect to genes with tissue specific functions. Additionally, we observe that genes with tissue-specific functions lie at the peripheries of our networks. We identify numerous modules enriched for Gene Ontology functions, and show that modules conserved across tissues are especially likely to have functions common to all tissues, while modules that are upregulated in a particular tissue are often instrumental to tissue-specific function. Finally, we provide a web tool, available at mostafavilab.stat.ubc.ca/GNAT, which allows exploration of gene function and regulation in a tissue-specific manner.