The large amount of time clinicians spend sifting through patient notes and documenting in electronic health records (EHRs) is a leading cause of clinician burnout.
We aimed to build machine learning algorithms to identify pregnant patients and triage them by risk of complication to assist care management.
In this position paper, we argue that developing AI supports for expository writing has unique and exciting research challenges and can lead to high real-world impacts.
We develop a predictive inference procedure that combines conformal prediction (CP) with unconditional quantile regression (QR) -- a commonly used tool in econometrics that involves regressing the recentered influence function (RIF) of the quantile functional over input covariates.
Our approach is interpretable, allowing a practitioner to visualize which subgroups in the population lead to falsification of an observational study.
We show that prior approaches can fail to find a human-AI system with low misclassification error even when there exists a linear classifier and rejector that have zero error (the realizable setting).
We study the application of large language models to zero-shot and few-shot classification of tabular data.
Under the assumption that at least one observational estimator is asymptotically normal and consistent for both the validation and extrapolated effects, we provide guarantees on the coverage probability of the intervals output by our algorithm.
One of the goals of learning algorithms is to complement and reduce the burden on human decision makers.
Subset selection applies to any label model and classifier and is very simple to plug in to existing weak supervision pipelines, requiring just a few lines of code.
We give a method for proactively identifying small, plausible shifts in distribution which lead to large differences in model performance.
A long-running goal of the clinical NLP community is the extraction of important variables trapped in clinical notes.
We demonstrate that co-training (Blum & Mitchell, 1998) can improve the performance of prompt-based learning by using unlabeled data.
For this collaboration to perform properly, the human decision maker must have a mental model of when and when not to rely on the agent.
Label-scarce, high-dimensional domains such as healthcare present a challenge for modern machine learning techniques.
Our question is when using this privileged data leads to more sample-efficient learning of models that use only baseline data for predictions at test time.
Individuals often make different decisions when faced with the same context, due to personal preferences and background.
In this work, we describe our creation of a dataset of clinical action items annotated over MIMIC-III, the largest publicly available dataset of real clinical notes.
Automated decision support can accelerate tedious tasks as users can focus their attention where it is needed most.
Decision Making Human-Computer Interaction
In the case of two proxy variables, we propose a modified estimator that is prediction optimal under interventions up to a known strength.
Several works have shown that perturbation stable instances of the MAP inference problem in Potts models can be solved exactly using a natural linear programming (LP) relaxation.
In this work, we focus on mitigating the interference of interval censoring in the task of clustering for disease phenotyping.
On "real-world" instances, MAP assignments of small perturbations of the problem should be very similar to the MAP assignment(s) of the original problem instance.
Reinforcement learning (RL) has the potential to significantly improve clinical decision making.
We reformulate the annotation framework for clinical entity extraction to factor in these issues to allow for robust end-to-end system benchmarking.
We present a system that uses a learned autocompletion mechanism to facilitate rapid creation of semi-structured clinical documentation.
Data cleaning is naturally framed as probabilistic inference in a generative model of ground-truth data and likely errors, but the diversity of real-world error patterns and the hardness of inference make Bayesian approaches difficult to automate.
Healthcare providers are increasingly using machine learning to predict patient outcomes to make meaningful interventions.
In several medical decision-making problems, such as antibiotic prescription, laboratory testing can provide precise indications for how a patient will respond to different treatment options.
Both electronic health records and personal health records are typically organized by data type, with medical problems, medications, procedures, and laboratory results chronologically sorted in separate areas of the chart.
Practitioners in diverse fields such as healthcare, economics and education are eager to apply machine learning to improve decision making.
Estimation of individual treatment effects is commonly used as the basis for contextual decision making in fields such as healthcare, education, and economics.
Machine-learned diagnosis models have shown promise as medical aides but are trained under a closed-set assumption, i. e. that models will only encounter conditions on which they have been trained.
Increasingly large electronic health records (EHRs) provide an opportunity to algorithmically learn medical knowledge.
One of the most surprising and exciting discoveries in supervising learning was the benefit of overparameterization (i. e. training a very large model) to improving the optimization landscape of a problem, with minimal effect on statistical performance (i. e. generalization).
Overlap between treatment groups is required for non-parametric estimation of causal effects.
One of the most surprising and exciting discoveries in supervised learning was the benefit of overparameterization (i. e. training a very large model) to improving the optimization landscape of a problem, with minimal effect on statistical performance (i. e. generalization).
We introduce an off-policy evaluation procedure for highlighting episodes where applying a reinforcement learned (RL) policy is likely to have produced a substantially different outcome than the observed policy.
In this work, we give generalization bounds for unsupervised domain adaptation that hold for any representation function by acknowledging the cost of non-invertibility.
Moreover, we conjecture that the proposed program recovers a mixing component at the rate k < p^2/4 and prove that a mixing component can be recovered with high probability when k < (2 - epsilon) p log p when the original components are sampled uniformly at random on the hyper sphere.
We consider the problem of image classification for the purpose of aiding doctors in dermatological diagnosis.
no code implementations • 31 May 2018 • Omer Gottesman, Fredrik Johansson, Joshua Meier, Jack Dent, Dong-hun Lee, Srivatsan Srinivasan, Linying Zhang, Yi Ding, David Wihl, Xuefeng Peng, Jiayu Yao, Isaac Lage, Christopher Mosch, Li-wei H. Lehman, Matthieu Komorowski, Aldo Faisal, Leo Anthony Celi, David Sontag, Finale Doshi-Velez
Much attention has been devoted recently to the development of machine learning algorithms with the goal of improving treatment policies in healthcare.
Amortized variational inference (AVI) replaces instance-specific local inference with a global inference network.
Ranked #2 on Text Generation on Yahoo Questions
Approximate algorithms for structured prediction problems---such as LP relaxations and the popular alpha-expansion algorithm (Boykov et al. 2001)---typically far exceed their theoretical performance guarantees on real-world instances.
Learning individual-level causal effects from observational data, such as inferring the most effective medication for a specific patient, is a problem of growing importance for policy makers.
Ranked #9 on Causal Inference on IHDP
In this work, we present the Grounded Recurrent Neural Network (GRNN), a recurrent neural network architecture for multi-label prediction which explicitly ties labels to specific dimensions of the recurrent hidden state (we call this process "grounding").
This work presents a novel objective function for the unsupervised training of neural network sentence encoders.
We consider multi-class classification where the predictor has a hierarchical structure that allows for a very large number of labels both at train and test time.
We introduce a unified algorithm to efficiently learn a broad class of linear and non-linear state space models, including variants where the emission and transition distributions are modeled by deep neural networks.
Ranked #4 on Multivariate Time Series Forecasting on USHCN-Daily
Disparate areas of machine learning have benefited from models that can take raw data with little preprocessing as input and learn rich representations of that raw data in order to perform well on a given prediction task.
We describe a method for parameter estimation in bipartite probabilistic graphical models for joint prediction of clinical conditions from the electronic medical record.
This work proposes a new algorithm for automated and simultaneous phenotyping of multiple co-occurring medical conditions, also referred as comorbidities, using clinical notes from the electronic health records (EHRs).
We give a novel, simple and intuitive generalization-error bound showing that the expected ITE estimation error of a representation is bounded by a sum of the standard generalization-error of that representation and the distance between the treated and control distributions induced by the representation.
Ranked #3 on Causal Inference on Jobs
Multivariate time series data in practical applications, such as health care, geoscience, and biology, are characterized by a variety of missing values.
Ranked #4 on Multivariate Time Series Forecasting on MuJoCo
Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology.
Early diagnosis of treatable diseases is essential for improving healthcare, and many diseases' onsets are predictable from annual lab tests and their temporal trends.
Motivated by recent variational methods for learning deep generative models, we introduce a unified algorithm to efficiently learn a broad spectrum of Kalman filters.
Structured prediction is used in areas such as computer vision and natural language processing to predict structured outputs such as segmentations or parse trees.
We show that the prospects for achieving low expected Hamming error depend on the structure of the graph $G$ in interesting ways.
We first show that for these graphical models, the tree-reweighted variational objective lends itself to a compact lifted formulation which can be solved much more efficiently than the standard TRW formulation for the ground graphical model.
We show that the existence of such a quartet allows us to uniquely identify each latent variable and to learn all parameters involving that latent variable.
This paper considers the problem of learning the parameters in Bayesian networks of discrete variables with known structure and hidden variables.
Topic models provide a useful method for dimensionality reduction and exploratory data analysis in large text corpora.
We propose a new class of consistency constraints for Linear Programming (LP) relaxations for finding the most probable (MAP) configuration in graphical models.