no code implementations • 15 Mar 2024 • Zhe Huang, Xiaowei Yu, Dajiang Zhu, Michael C. Hughes
In this paper, we introduce InterLUDE, a new approach to enhance SSL made of two parts that each benefit from labeled-unlabeled interaction.
no code implementations • 9 Mar 2024 • Zhe Huang, Xiaowei Yu, Benjamin S. Wessler, Michael C. Hughes
However, existing deep learning pipelines for assessing AS from echocardiograms have two key limitations.
no code implementations • 26 Jan 2024 • Michael Wojnowicz, Preetish Rath, Eric Miller, Jeffrey Miller, Clifford Hancock, Meghan O'Donovan, Seth Elkin-Frankston, Thaddeus Brunye, Michael C. Hughes
Our hierarchical switching recurrent dynamical models can be learned via closed-form variational coordinate ascent updates to all latent chains that scale linearly in the number of individual time series.
1 code implementation • 29 Nov 2023 • Ethan Harvey, Wansu Chen, David M. Kent, Michael C. Hughes
Practitioners building classifiers often start with a smaller pilot dataset and plan to grow to larger data in the near future.
1 code implementation • 25 Sep 2023 • Patrick Feeney, Michael C. Hughes
The information noise-contrastive estimation (InfoNCE) loss function provides the basis of many self-supervised deep learning methods due to its strong empirical results and theoretic motivation.
1 code implementation • 18 Jul 2023 • Zhe Huang, Ruijie Jiang, Shuchin Aeron, Michael C. Hughes
Yet past benchmarks do not focus on medical tasks and rarely compare self- and semi- methods together on an equal footing.
1 code implementation • 25 May 2023 • Zhe Huang, Benjamin S. Wessler, Michael C. Hughes
To automate screening for AS, deep networks must learn to mimic a human expert's ability to identify views of the aortic valve then aggregate across these relevant images to produce a study-level diagnosis.
no code implementations • 4 Oct 2022 • Kevin C. Cheng, Shuchin Aeron, Michael C. Hughes, Eric L. Miller
We consider probabilistic models for sequential observations which exhibit gradual transitions among a finite number of states.
1 code implementation • 25 Aug 2022 • Zhe Huang, Mary-Joy Sidhom, Benjamin S. Wessler, Michael C. Hughes
Semi-supervised learning (SSL) promises improved accuracy compared to training classifiers on small labeled datasets by also training on many unlabeled images.
2 code implementations • 23 Jun 2022 • Patrick Feeney, Sarah Schneider, Panagiotis Lymperopoulos, Li-Ping Liu, Matthias Scheutz, Michael C. Hughes
In order for artificial agents to successfully perform tasks in changing environments, they must be able to both detect and adapt to novelty.
1 code implementation • 31 May 2022 • Michael T. Wojnowicz, Shuchin Aeron, Eric L. Miller, Michael C. Hughes
This approximation makes inference straightforward and fast; using well-known auxiliary variables for probit or logistic regression, the product of binary models admits conjugate closed-form variational inference that is embarrassingly parallel across categories and invariant to category ordering.
1 code implementation • NeurIPS 2021 • Kevin C. Cheng, Shuchin Aeron, Michael C. Hughes, Eric L. Miller
We propose a dynamical Wasserstein barycentric (DWB) model that estimates the system state over time as well as the data-generating distributions of pure states in an unsupervised manner.
1 code implementation • 30 Jul 2021 • Zhe Huang, Gary Long, Benjamin Wessler, Michael C. Hughes
Semi-supervised image classification has shown substantial progress in learning from limited labeled data, but recent advances remain largely untested for clinical applications.
no code implementations • 28 Jul 2021 • Patrick Feeney, Michael C. Hughes
The pixelwise reconstruction error of deep autoencoders is often utilized for image novelty detection and localization under the assumption that pixels with high error indicate which parts of the input image are unfamiliar and therefore likely to be novel.
1 code implementation • 4 Jun 2021 • Linfeng Liu, Michael C. Hughes, Soha Hassoun, Li-Ping Liu
In this work, we propose a new model, Stochastic Iterative Graph MAtching (SIGMA), to address the graph matching problem.
1 code implementation • 28 Apr 2021 • Gian Marco Visani, Alexandra Hope Lee, Cuong Nguyen, David M. Kent, John B. Wong, Joshua T. Cohen, Michael C. Hughes
We develop an Approximate Bayesian Computation approach that draws samples from the posterior distribution over the model's transition and duration parameters given aggregate counts from a specific location, thus adapting the model to a region or individual hospital site of interest.
1 code implementation • 14 Apr 2021 • Alexandra Hope Lee, Panagiotis Lymperopoulos, Joshua T. Cohen, John B. Wong, Michael C. Hughes
We consider the problem of forecasting the daily number of hospitalized COVID-19 patients at a single hospital site, in order to help administrators with logistics and planning.
no code implementations • 29 Mar 2021 • Linfeng Liu, Michael C. Hughes, Li-Ping Liu
We propose a new model, the Neighbor Mixture Model (NMM), for modeling node labels in a graph.
no code implementations • 12 Dec 2020 • Gabriel Hope, Madina Abdrakhmanova, Xiaoyin Chen, Michael C. Hughes, Erik B. Sudderth
We develop a new framework for learning variational autoencoders and other deep generative models that balances generative and discriminative goals.
no code implementations • 9 Jun 2020 • Kevin C. Cheng, Eric L. Miller, Michael C. Hughes, Shuchin Aeron
Non-parametric and distribution-free two-sample tests have been the foundation of many change point detection algorithms.
1 code implementation • 18 Feb 2020 • Gian Marco Visani, Michael C. Hughes, Soha Hassoun
Some interactions are attributed to natural selection and involve the enzyme's natural substrates.
1 code implementation • 13 Jan 2020 • Joseph Futoma, Michael C. Hughes, Finale Doshi-Velez
Many medical decision-making tasks can be framed as partially observed Markov decision processes (POMDPs).
no code implementations • 4 Nov 2019 • Kevin C. Cheng, Shuchin Aeron, Michael C. Hughes, Erika Hussey, Eric L. Miller
Two common problems in time series analysis are the decomposition of the data stream into disjoint segments that are each in some sense "homogeneous" - a problem known as Change Point Detection (CPD) - and the grouping of similar nonadjacent segments, a problem that we call Time Series Segment Clustering (TSSC).
no code implementations • pproximateinference AABI Symposium 2019 • Melanie F. Pradier, Michael C. Hughes, Finale Doshi-Velez
Variational inference based on chi-square divergence minimization (CHIVI) provides a way to approximate a model's posterior while obtaining an upper bound on the marginal likelihood.
1 code implementation • pproximateinference AABI Symposium 2019 • Lily H. Zhang, Michael C. Hughes
Comparing the inferences of diverse candidate models is an essential part of model checking and escaping local optima.
no code implementations • 14 Aug 2019 • Mike Wu, Sonali Parbhoo, Michael C. Hughes, Volker Roth, Finale Doshi-Velez
Moreover, for situations in which a single, global tree is a poor estimator, we introduce a regional tree regularizer that encourages the deep model to resemble a compact, axis-aligned decision tree in predefined, human-interpretable contexts.
1 code implementation • 2 Aug 2019 • Bret Nestor, Matthew B. A. McDermott, Willie Boag, Gabriela Berner, Tristan Naumann, Michael C. Hughes, Anna Goldenberg, Marzyeh Ghassemi
When training clinical prediction models from electronic health records (EHRs), a key concern should be a model's ability to sustain performance over time when deployed, even as care practices, database systems, and population demographics evolve.
2 code implementations • 19 Jul 2019 • Shirly Wang, Matthew B. A. McDermott, Geeticka Chauhan, Michael C. Hughes, Tristan Naumann, Marzyeh Ghassemi
Robust machine learning relies on access to data that can be used with standardized frameworks in important tasks and the ability to develop models whose performance can be reasonably reproduced.
Ranked #3 on Length-of-Stay prediction on MIMIC-III
no code implementations • 30 Nov 2018 • Bret Nestor, Matthew B. A. McDermott, Geeticka Chauhan, Tristan Naumann, Michael C. Hughes, Anna Goldenberg, Marzyeh Ghassemi
Machine learning for healthcare often trains models on de-identified datasets with randomly-shifted calendar dates, ignoring the fact that data were generated under hospital operation practices that change over time.
no code implementations • 1 Dec 2017 • Michael C. Hughes, Gabriel Hope, Leah Weiner, Thomas H. McCoy, Roy H. Perlis, Erik B. Sudderth, Finale Doshi-Velez
Supervisory signals can help topic models discover low-dimensional data representations that are more interpretable for clinical tasks.
2 code implementations • 16 Nov 2017 • Mike Wu, Michael C. Hughes, Sonali Parbhoo, Maurizio Zazzi, Volker Roth, Finale Doshi-Velez
The lack of interpretability remains a key barrier to the adoption of deep models in many applications.
1 code implementation • ICML 2017 • Geng Ji, Michael C. Hughes, Erik B. Sudderth
Our model is based on a novel, variational interpretation of the popular expected patch log-likelihood (EPLL) method as a model for randomly positioned grids of image patches.
no code implementations • 23 Jul 2017 • Michael C. Hughes, Leah Weiner, Gabriel Hope, Thomas H. McCoy Jr., Roy H. Perlis, Erik B. Sudderth, Finale Doshi-Velez
Supervisory signals have the potential to make low-dimensional data representations, like those learned by mixture and topic models, more interpretable and useful.
1 code implementation • 10 Mar 2017 • Andrew Slavin Ross, Michael C. Hughes, Finale Doshi-Velez
Neural networks are among the most accurate supervised learning methods in use today, but their opacity makes them difficult to trust in critical applications, especially when conditions in training differ from those in test.
no code implementations • 6 Dec 2016 • Michael C. Hughes, Huseyin Melih Elibol, Thomas McCoy, Roy Perlis, Finale Doshi-Velez
Supervised topic models can help clinical researchers find interpretable cooccurence patterns in count data that are relevant for diagnostics.
no code implementations • 23 Sep 2016 • Michael C. Hughes, Erik B. Sudderth
Mixture models and topic models generate each observation from a single cluster, but standard variational posteriors for each observation assign positive probability to all possible clusters.
1 code implementation • NeurIPS 2015 • Michael C. Hughes, William T. Stephenson, Erik Sudderth
Bayesian nonparametric hidden Markov models are typically learned via fixed truncations of the infinite state space or local Monte Carlo proposals that make small changes to the state space.
no code implementations • NeurIPS 2013 • Michael C. Hughes, Erik Sudderth
Variational inference algorithms provide the most effective framework for large-scale training of Bayesian nonparametric models.
no code implementations • 22 Aug 2013 • Emily B. Fox, Michael C. Hughes, Erik B. Sudderth, Michael. I. Jordan
We propose a Bayesian nonparametric approach to the problem of jointly modeling multiple related time series.
no code implementations • NeurIPS 2012 • Michael C. Hughes, Emily Fox, Erik B. Sudderth
Applications of Bayesian nonparametric methods require learning and inference algorithms which efficiently explore models of unbounded complexity.