Search Results for author: Mihaela van der Schaar

Found 140 papers, 56 papers with code

Disentangled Counterfactual Recurrent Networks for Treatment Effect Inference over Time

no code implementations7 Dec 2021 Jeroen Berrevoets, Alicia Curth, Ioana Bica, Eoin McKinney, Mihaela van der Schaar

Choosing the best treatment-plan for each individual patient requires accurate forecasts of their outcome trajectories as a function of the treatment, over time.

SyncTwin: Treatment Effect Estimation with Longitudinal Outcomes

1 code implementation NeurIPS 2021 Zhaozhi Qian, Yao Zhang, Ioana Bica, Angela Wood, Mihaela van der Schaar

Most of the medical observational studies estimate the causal treatment effects using electronic health records (EHR), where a patient's covariates and outcomes are both observed longitudinally.

Invariant Causal Imitation Learning for Generalizable Policies

no code implementations NeurIPS 2021 Ioana Bica, Daniel Jarrett, Mihaela van der Schaar

By leveraging data from multiple environments, we propose Invariant Causal Imitation Learning (ICIL), a novel technique in which we learn a feature representation that is invariant across domains, on the basis of which we learn an imitation policy that matches expert behavior.

Imitation Learning

Estimating Multi-cause Treatment Effects via Single-cause Perturbation

1 code implementation NeurIPS 2021 Zhaozhi Qian, Alicia Curth, Mihaela van der Schaar

Most existing methods for conditional average treatment effect estimation are designed to estimate the effect of a single cause - only one variable can be intervened on at one time.

Causal Inference

Conformal Time-series Forecasting

1 code implementation NeurIPS 2021 Kamile Stankeviciute, Ahmed M. Alaa, Mihaela van der Schaar

Current approaches for multi-horizon time series forecasting using recurrent neural networks (RNNs) focus on issuing point estimates, which is insufficient for decision-making in critical application domains where an uncertainty estimate is also required.

Decision Making Time Series +1

Time-series Generation by Contrastive Imitation

no code implementations NeurIPS 2021 Daniel Jarrett, Ioana Bica, Mihaela van der Schaar

In this work, we study a generative framework that seeks to combine the strengths of both: Motivated by a moment-matching objective to mitigate compounding error, we optimize a local (but forward-looking) transition policy, where the reinforcement signal is provided by a global (but stepwise-decomposable) energy model trained by contrastive estimation.

Time Series

Explaining Latent Representations with a Corpus of Examples

1 code implementation NeurIPS 2021 Jonathan Crabbé, Zhaozhi Qian, Fergus Imrie, Mihaela van der Schaar

SimplEx uses the corpus to improve the user's understanding of the latent space with post-hoc explanations answering two questions: (1) Which corpus examples explain the prediction issued for a given test example?

Image Classification Mortality Prediction

Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

no code implementations8 Sep 2021 Mahed Abroshan, Kai Hou Yip, Cem Tekin, Mihaela van der Schaar

Secondly, such datasets are usually imperfect, additionally cursed with missing values in the attributes of features.

Decision Making

Identifiable Energy-based Representations: An Application to Estimating Heterogeneous Causal Effects

1 code implementation6 Aug 2021 Yao Zhang, Jeroen Berrevoets, Mihaela van der Schaar

Conditional average treatment effects (CATEs) allow us to understand the effect heterogeneity across a large population of individuals.

Dimensionality Reduction

Doing Great at Estimating CATE? On the Neglected Assumptions in Benchmark Comparisons of Treatment Effect Estimators

no code implementations28 Jul 2021 Alicia Curth, Mihaela van der Schaar

The machine learning toolbox for estimation of heterogeneous treatment effects from observational data is expanding rapidly, yet many of its algorithms have been evaluated only on a very limited set of semi-synthetic benchmark datasets.

Inverse Contextual Bandits: Learning How Behavior Evolves over Time

no code implementations13 Jul 2021 Alihan Hüyük, Daniel Jarrett, Mihaela van der Schaar

Understanding a decision-maker's priorities by observing their behavior is critical for transparency and accountability in decision processes, such as in healthcare.

Decision Making Multi-Armed Bandits

The Medkit-Learn(ing) Environment: Medical Decision Modelling through Simulation

1 code implementation8 Jun 2021 Alex J. Chan, Ioana Bica, Alihan Huyuk, Daniel Jarrett, Mihaela van der Schaar

Understanding decision-making in clinical environments is of paramount importance if we are to bring the strengths of machine learning to ultimately improve patient outcomes.

Decision Making

On Inductive Biases for Heterogeneous Treatment Effect Estimation

1 code implementation NeurIPS 2021 Alicia Curth, Mihaela van der Schaar

We investigate how to exploit structural similarities of an individual's potential outcomes (POs) under different treatments to obtain better estimates of conditional average treatment effects in finite samples.

POS

Integrating Expert ODEs into Neural ODEs: Pharmacology and Disease Progression

no code implementations NeurIPS 2021 Zhaozhi Qian, William R. Zame, Lucas M. Fleuren, Paul Elbers, Mihaela van der Schaar

To close this gap, we propose the latent hybridisation model (LHM) that integrates a system of expert-designed ODEs with machine-learned Neural ODEs to fully describe the dynamics of the system and to link the expert and latent variables to observable quantities.

Graphical modelling in continuous-time: consistency guarantees and algorithms using Neural ODEs

no code implementations6 May 2021 Alexis Bellot, Kim Branson, Mihaela van der Schaar

Using this insight, we propose a score-based learning algorithm based on penalized Neural Ordinary Differential Equations (modelling the mean process) that we show to be applicable to the general setting of irregularly-sampled multivariate time series and to outperform the state of the art across a range of dynamical systems.

Causal Discovery Graph Learning +1

Model-Attentive Ensemble Learning for Sequence Modeling

no code implementations23 Feb 2021 Victor D. Bourgin, Ioana Bica, Mihaela van der Schaar

Medical time-series datasets have unique characteristics that make prediction tasks challenging.

Ensemble Learning Time Series

How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models

no code implementations17 Feb 2021 Ahmed M. Alaa, Boris van Breugel, Evgeny Saveliev, Mihaela van der Schaar

In this paper, we introduce a 3-dimensional evaluation metric, ($\alpha$-Precision, $\beta$-Recall, Authenticity), that characterizes the fidelity, diversity and generalization performance of any generative model in a domain-agnostic fashion.

Image Generation

Scalable Bayesian Inverse Reinforcement Learning

2 code implementations12 Feb 2021 Alex J. Chan, Mihaela van der Schaar

Bayesian inference over the reward presents an ideal solution to the ill-posed nature of the inverse reinforcement learning problem.

Bayesian Inference Imitation Learning

Selecting Treatment Effects Models for Domain Adaptation Using Causal Knowledge

no code implementations11 Feb 2021 Trent Kyono, Ioana Bica, Zhaozhi Qian, Mihaela van der Schaar

We leverage the invariance of causal structures across domains to propose a novel model selection metric specifically designed for ITE methods under the UDA setting.

Causal Inference Model Selection +1

A Variational Information Bottleneck Approach to Multi-Omics Data Integration

1 code implementation5 Feb 2021 Changhee Lee, Mihaela van der Schaar

Due to non-uniformity and technical limitations in omics platforms, such integrative analyses on multiple omics, which we refer to as views, involve learning from incomplete observations with various view-missing patterns.

Policy Analysis using Synthetic Controls in Continuous-Time

1 code implementation2 Feb 2021 Alexis Bellot, Mihaela van der Schaar

Counterfactual estimation using synthetic controls is one of the most successful recent methodological developments in causal inference.

Causal Inference Time Series

Learning Matching Representations for Individualized Organ Transplantation Allocation

1 code implementation28 Jan 2021 Can Xu, Ahmed M. Alaa, Ioana Bica, Brent D. Ershoff, Maxime Cannesson, Mihaela van der Schaar

Organ transplantation is often the last resort for treating end-stage illness, but the probability of a successful transplantation depends greatly on compatibility between donors and recipients.

Representation Learning

SDF-Bayes: Cautious Optimism in Safe Dose-Finding Clinical Trials with Drug Combinations and Heterogeneous Patient Groups

no code implementations26 Jan 2021 Hyun-Suk Lee, Cong Shen, William Zame, Jang-Won Lee, Mihaela van der Schaar

Phase I clinical trials are designed to test the safety (non-toxicity) of drugs and find the maximum tolerated dose (MTD).

Nonparametric Estimation of Heterogeneous Treatment Effects: From Theory to Learning Algorithms

2 code implementations26 Jan 2021 Alicia Curth, Mihaela van der Schaar

The need to evaluate treatment effectiveness is ubiquitous in most of empirical science, and interest in flexibly investigating effect heterogeneity is growing rapidly.

Meta-Learning

Personalized Education in the AI Era: What to Expect Next?

no code implementations19 Jan 2021 Setareh Maghsudi, Andrew Lan, Jie Xu, Mihaela van der Schaar

The objective of personalized learning is to design an effective knowledge acquisition track that matches the learner's strengths and bypasses her weaknesses to ultimately meet her desired goal.

Scalable Bayesian Inverse Reinforcement Learning by Auto-Encoding Reward

no code implementations ICLR 2021 Alex James Chan, Mihaela van der Schaar

Bayesian inference over the reward presents an ideal solution to the ill-posed nature of the inverse reinforcement learning problem.

Bayesian Inference Imitation Learning

SyncTwin: Transparent Treatment Effect Estimation under Temporal Confounding

no code implementations1 Jan 2021 Zhaozhi Qian, Yao Zhang, Ioana Bica, Angela Wood, Mihaela van der Schaar

Estimating causal treatment effects using observational data is a problem with few solutions when the confounder has a temporal structure, e. g. the history of disease progression might impact both treatment decisions and clinical outcomes.

Clairvoyance: A Pipeline Toolkit for Medical Time Series

no code implementations ICLR 2021 Daniel Jarrett, Jinsung Yoon, Ioana Bica, Zhaozhi Qian, Ari Ercole, Mihaela van der Schaar

Despite exponential growth in electronic patient data, there is a remarkable gap between the potential and realized utilization of ML for clinical research and decision support.

AutoML Time Series

Generative Time-series Modeling with Fourier Flows

no code implementations ICLR 2021 Ahmed Alaa, Alex James Chan, Mihaela van der Schaar

Generating synthetic time-series data is crucial in various application domains, such as medical prognosis, wherein research is hamstrung by the lack of access to data due to concerns over privacy.

Time Series

Synthetic Data: Opening the data floodgates to enable faster, more directed development of machine learning methods

no code implementations8 Dec 2020 James Jordon, Alan Wilson, Mihaela van der Schaar

Many ground-breaking advancements in machine learning can be attributed to the availability of a large volume of rich data.

Gradient Regularized V-Learning for Dynamic Treatment Regimes

1 code implementation NeurIPS 2020 Yao Zhang, Mihaela van der Schaar

A dynamic treatment regime (DTR) is a sequence of treatment rules indicating how to individualize treatments for a patient based on the previously assigned treatments and the evolving covariate history.

Learning outside the Black-Box: The pursuit of interpretable models

1 code implementation NeurIPS 2020 Jonathan Crabbé, Yao Zhang, William Zame, Mihaela van der Schaar

Machine Learning has proved its ability to produce accurate models but the deployment of these models outside the machine learning community has been hindered by the difficulties of interpreting these models.

Estimating Structural Target Functions using Machine Learning and Influence Functions

1 code implementation14 Aug 2020 Alicia Curth, Ahmed M. Alaa, Mihaela van der Schaar

Within this framework, we propose two general learning algorithms that build on the idea of nonparametric plug-in bias removal via IFs: the 'IF-learner' which uses pseudo-outcomes motivated by uncentered IFs for regression in large samples and outputs entire target functions without confidence bands, and the 'Group-IF-learner', which outputs only approximations to a function but can give confidence estimates if sufficient information on coarsening mechanisms is available.

Epidemiology

CPAS: the UK's National Machine Learning-based Hospital Capacity Planning System for COVID-19

no code implementations27 Jul 2020 Zhaozhi Qian, Ahmed M. Alaa, Mihaela van der Schaar

The coronavirus disease 2019 (COVID-19) global pandemic poses the threat of overwhelming healthcare systems with unprecedented demands for intensive care resources.

Accounting for Unobserved Confounding in Domain Generalization

no code implementations21 Jul 2020 Alexis Bellot, Mihaela van der Schaar

The ability to generalize from observed to new related environments is central to any form of reliable machine learning, yet most methods fail when moving beyond i. i. d data.

Domain Generalization

Learning "What-if" Explanations for Sequential Decision-Making

no code implementations ICLR 2021 Ioana Bica, Daniel Jarrett, Alihan Hüyük, Mihaela van der Schaar

Building interpretable parameterizations of real-world decision-making on the basis of demonstrated behavior -- i. e. trajectories of observations and actions made by an expert maximizing some unknown reward function -- is essential for introspecting and auditing policies in different institutions.

Decision Making

Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functions

1 code implementation ICML 2020 Ahmed M. Alaa, Mihaela van der Schaar

Deep learning models achieve high predictive accuracy across a broad spectrum of tasks, but rigorously quantifying their predictive uncertainty remains challenging.

Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift

no code implementations ICML 2020 Alex J. Chan, Ahmed M. Alaa, Zhaozhi Qian, Mihaela van der Schaar

In this paper, we develop an approximate Bayesian inference scheme based on posterior regularisation, wherein unlabelled target data are used as "pseudo-labels" of model confidence that are used to regularise the model's loss on labelled source data.

Bayesian Inference Decision Making

Inverse Active Sensing: Modeling and Understanding Timely Decision-Making

no code implementations ICML 2020 Daniel Jarrett, Mihaela van der Schaar

Finally, we illustrate how this formulation enables understanding decision-making behavior by quantifying preferences implicit in observed decision strategies (the inverse problem).

Decision Making

Strictly Batch Imitation Learning by Energy-based Distribution Matching

1 code implementation NeurIPS 2020 Daniel Jarrett, Ioana Bica, Mihaela van der Schaar

Through experiments with application to control and healthcare settings, we illustrate consistent performance gains over existing algorithms for strictly batch imitation learning.

Imitation Learning

AutoCP: Automated Pipelines for Accurate Prediction Intervals

no code implementations24 Jun 2020 Yao Zhang, William Zame, Mihaela van der Schaar

Successful application of machine learning models to real-world prediction problems, e. g. financial forecasting and personalized medicine, has proved to be challenging, because such settings require limiting and quantifying the uncertainty in the model predictions, i. e. providing valid and accurate prediction intervals.

AutoML Prediction Intervals +1

Temporal Phenotyping using Deep Predictive Clustering of Disease Progression

1 code implementation ICML 2020 Changhee Lee, Mihaela van der Schaar

In this paper, we develop a deep learning approach for clustering time-series data, where each cluster comprises patients who share similar future outcomes of interest (e. g., adverse events, the onset of comorbidities).

Decision Making Patient Phenotyping +2

Robust Recursive Partitioning for Heterogeneous Treatment Effects with Uncertainty Quantification

1 code implementation NeurIPS 2020 Hyun-Suk Lee, Yao Zhang, William Zame, Cong Shen, Jang-Won Lee, Mihaela van der Schaar

Most of the current methods of subgroup analysis begin with a particular algorithm for estimating individualized treatment effects (ITE) and identify subgroups by maximizing the difference across subgroups of the average treatment effect in each subgroup.

Recommendation Systems

Learning for Dose Allocation in Adaptive Clinical Trials with Safety Constraints

no code implementations ICML 2020 Cong Shen, Zhiyang Wang, Sofia S. Villar, Mihaela van der Schaar

Phase I dose-finding trials are increasingly challenging as the relationship between efficacy and toxicity of new compounds (or combination of them) becomes more complex.

When and How to Lift the Lockdown? Global COVID-19 Scenario Analysis and Policy Assessment using Compartmental Gaussian Processes

1 code implementation NeurIPS 2020 Zhaozhi Qian, Ahmed M. Alaa, Mihaela van der Schaar

To this end, this paper develops a Bayesian model for predicting the effects of COVID-19 lockdown policies in a global context -- we treat each country as a distinct data point, and exploit variations of policies across countries to learn country-specific policy effects.

Gaussian Processes Variational Inference

A Non-Stationary Bandit-Learning Approach to Energy-Efficient Femto-Caching with Rateless-Coded Transmission

no code implementations13 Apr 2020 Setareh Maghsudi, Mihaela van der Schaar

The former problem boils down to a stochastic knapsack problem, and we cast the latter as a multi-armed bandit problem.

Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networks

1 code implementation NeurIPS 2020 Ioana Bica, James Jordon, Mihaela van der Schaar

While much attention has been given to the problem of estimating the effect of discrete interventions from observational data, relatively little work has been done in the setting of continuous-valued interventions, such as treatments associated with a dosage parameter.

Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations

2 code implementations ICLR 2020 Ioana Bica, Ahmed M. Alaa, James Jordon, Mihaela van der Schaar

Identifying when to give treatments to patients and how to select among multiple treatments over time are important medical problems with a few existing solutions.

Target-Embedding Autoencoders for Supervised Representation Learning

no code implementations ICLR 2020 Daniel Jarrett, Mihaela van der Schaar

Autoencoder-based learning has emerged as a staple for disciplining representations in unsupervised and semi-supervised settings.

Representation Learning

Stepwise Model Selection for Sequence Prediction via Deep Kernel Learning

1 code implementation12 Jan 2020 Yao Zhang, Daniel Jarrett, Mihaela van der Schaar

In this paper, we propose a novel Bayesian optimization (BO) algorithm to tackle the challenge of model selection in this setting.

AutoML Model Selection

Contextual Constrained Learning for Dose-Finding Clinical Trials

1 code implementation8 Jan 2020 Hyun-Suk Lee, Cong Shen, James Jordon, Mihaela van der Schaar

In addition, patient recruitment can be difficult by the fact that clinical trials do not aim to provide a benefit to any given patient in the trial.

A Bayesian Approach to Modelling Longitudinal Data in Electronic Health Records

no code implementations19 Dec 2019 Alexis Bellot, Mihaela van der Schaar

Analyzing electronic health records (EHR) poses significant challenges because often few samples are available describing a patient's health and, when available, their information content is highly diverse.

Time-series Generative Adversarial Networks

1 code implementation1 Dec 2019 Jinsung Yoon, Daniel Jarrett, Mihaela van der Schaar

A good generative model for time-series data should preservetemporal dynamics, in the sense that new sequences respect the original relationships between variablesacross time.

Time Series

Differentially Private Bagging: Improved utility and cheaper privacy than subsample-and-aggregate

1 code implementation NeurIPS 2019 James Jordon, Jinsung Yoon, Mihaela van der Schaar

The second benefit is that, through analysis that we provide inthe paper, we can derive tighter differential privacy guarantees when several queriesare made to this mechanism.

Attentive State-Space Modeling of Disease Progression

1 code implementation NeurIPS 2019 Ahmed M. Alaa, Mihaela van der Schaar

Models of disease progression are instrumental for predicting patient outcomes and understanding disease dynamics.

Predicting Patient Outcomes

Demystifying Black-box Models with Symbolic Metamodels

1 code implementation NeurIPS 2019 Ahmed M. Alaa, Mihaela van der Schaar

A symbolic metamodel is a model of a model, i. e., a surrogate model of a trained (machine learning) model expressed through a succinct symbolic expression that comprises familiar mathematical functions and can be subjected to symbolic manipulation.

Improving Model Robustness Using Causal Knowledge

no code implementations27 Nov 2019 Trent Kyono, Mihaela van der Schaar

We show in this paper how prior knowledge in the form of a causal graph can be utilized to guide model selection, i. e., to identify from a set of trained networks the models that are the most robust and invariant to unseen domains.

Model Selection

The Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functions

no code implementations25 Sep 2019 Ahmed M. Alaa, Mihaela van der Schaar

To address this question, we develop the discriminative jackknife (DJ), a formal inference procedure that constructs predictive confidence intervals for a wide range of deep learning models, is easy to implement, and provides rigorous theoretical guarantees on (1) and (2).

Actor-Critic Approach for Temporal Predictive Clustering

no code implementations25 Sep 2019 Changhee Lee, Mihaela van der Schaar

In this paper, we develop a deep learning approach for clustering time-series data, where each cluster comprises patients who share similar future outcomes of interest (e. g., adverse events, the onset of comorbidities, etc.).

Decision Making Patient Phenotyping +2

Conditional Independence Testing using Generative Adversarial Networks

1 code implementation NeurIPS 2019 Alexis Bellot, Mihaela van der Schaar

We consider the hypothesis testing problem of detecting conditional dependence, with a focus on high-dimensional feature spaces.

Two-sample testing

Kernel Hypothesis Testing with Set-valued Data

no code implementations9 Jul 2019 Alexis Bellot, Mihaela van der Schaar

We present a general framework for hypothesis testing on distributions of sets of individual examples.

Time Series Two-sample testing

ASAC: Active Sensing using Actor-Critic models

1 code implementation16 Jun 2019 Jinsung Yoon, James Jordon, Mihaela van der Schaar

The predictor network uses the observations selected by the selector network to predict a label, providing feedback to the selector network (well-selected variables should be predictive of the label).

Lifelong Bayesian Optimization

no code implementations29 May 2019 Yao Zhang, James Jordon, Ahmed M. Alaa, Mihaela van der Schaar

In this paper, we present Lifelong Bayesian Optimization (LBO), an online, multitask Bayesian optimization (BO) algorithm designed to solve the problem of model selection for datasets arriving and evolving over time.

Model Selection

KnockoffGAN: Generating Knockoffs for Feature Selection using Generative Adversarial Networks

1 code implementation ICLR 2019 James Jordon, Jinsung Yoon, Mihaela van der Schaar

We demonstrate the capability of our model to perform feature selection, showing that it performs as well as the originally proposed knockoff generation model in the Gaussian setting and that it outperforms the original model in non-Gaussian settings, including on a real-world dataset.

A Deep Learning Approach for Dynamic Survival Analysis with Competing Risks

no code implementations ICLR 2019 Changhee Lee, Mihaela van der Schaar

Currently available survival analysis methods are limited in their ability to deal with complex, heterogeneous, and longitudinal data such as that available in primary care records, or in their ability to deal with multiple competing risks.

Survival Analysis

PATE-GAN: Generating Synthetic Data with Differential Privacy Guarantees

1 code implementation ICLR 2019 Jinsung Yoon, James Jordon, Mihaela van der Schaar

Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available.

Synthetic Data Generation

INVASE: Instance-wise Variable Selection using Neural Networks

1 code implementation ICLR 2019 Jinsung Yoon, James Jordon, Mihaela van der Schaar

The advent of big data brings with it data with more and more dimensions and thus a growing need to be able to efficiently select which features to use for a variety of problems.

Variable Selection

Machine Learning in the Air

no code implementations28 Apr 2019 Deniz Gunduz, Paul de Kerret, Nicholas D. Sidiropoulos, David Gesbert, Chandra Murthy, Mihaela van der Schaar

Thanks to the recent advances in processing speed and data acquisition and storage, machine learning (ML) is penetrating every facet of our lives, and transforming research in many areas in a fundamental manner.

Multitask Boosting for Survival Analysis with Competing Risks

no code implementations NeurIPS 2018 Alexis Bellot, Mihaela van der Schaar

The co-occurrence of multiple diseases among the general population is an important problem as those patients have more risk of complications and represent a large share of health care expenditure.

Survival Analysis

What is Interpretable? Using Machine Learning to Design Interpretable Decision-Support Systems

no code implementations27 Nov 2018 Owen Lahav, Nicholas Mastronarde, Mihaela van der Schaar

Our results demonstrate that ML experts cannot accurately predict which system outputs will maximize clinicians' confidence in the underlying neural network model, and suggest additional findings that have broad implications to the future of research into ML interpretability and the use of ML in medicine.

MATCH-Net: Dynamic Prediction in Survival Analysis using Convolutional Neural Networks

no code implementations26 Nov 2018 Daniel Jarrett, Jinsung Yoon, Mihaela van der Schaar

Accurate prediction of disease trajectories is critical for early identification and timely treatment of patients at risk.

Survival Analysis

Feature Selection for Survival Analysis with Competing Risks using Deep Learning

1 code implementation22 Nov 2018 Carl Rietschel, Jinsung Yoon, Mihaela van der Schaar

Deep learning models for survival analysis have gained significant attention in the literature, but they suffer from severe performance deficits when the dataset contains many irrelevant features.

Survival Analysis

Estimation of Individual Treatment Effect in Latent Confounder Models via Adversarial Learning

no code implementations21 Nov 2018 Changhee Lee, Nicholas Mastronarde, Mihaela van der Schaar

Estimating the individual treatment effect (ITE) from observational data is essential in medicine.

Risk-Stratify: Confident Stratification Of Patients Based On Risk

no code implementations2 Nov 2018 Kartik Ahuja, Mihaela van der Schaar

A clinician desires to use a risk-stratification method that achieves confident risk-stratification - the risk estimates of the different patients reflect the true risks with a high probability.

Joint Concordance Index

1 code implementation26 Oct 2018 Kartik Ahuja, Mihaela van der Schaar

We use the new metric to develop a variable importance ranking approach.

Methodology

Forecasting Individualized Disease Trajectories using Interpretable Deep Learning

no code implementations24 Oct 2018 Ahmed M. Alaa, Mihaela van der Schaar

In this paper, we develop the phased attentive state space (PASS) model of disease progression, a deep probabilistic model that captures complex representations for disease progression while maintaining clinical interpretability.

Disease Trajectory Forecasting

Adaptive Clinical Trials: Exploiting Sequential Patient Recruitment and Allocation

no code implementations5 Oct 2018 Onur Atan, William R. Zame, Mihaela van der Schaar

Randomized Controlled Trials (RCTs) are the gold standard for comparing the effectiveness of a new treatment to the current one (the control).

Siamese Survival Analysis with Competing Risks

no code implementations ICLR 2018 Anton Nemchenko, Trent Kyono, Mihaela van der Schaar

Survival analysis in the presence of multiple possible adverse events, i. e., competing risks, is a pervasive problem in many industries (healthcare, finance, etc.).

Survival Analysis

Forecasting Disease Trajectories in Alzheimer's Disease Using Deep Learning

no code implementations6 Jul 2018 Bryan Lim, Mihaela van der Schaar

Joint models for longitudinal and time-to-event data are commonly used in longitudinal studies to forecast disease trajectories over time.

Measuring the quality of Synthetic data for use in competitions

1 code implementation29 Jun 2018 James Jordon, Jinsung Yoon, Mihaela van der Schaar

Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available.

Optimal Piecewise Local-Linear Approximations

1 code implementation27 Jun 2018 Kartik Ahuja, William Zame, Mihaela van der Schaar

Piecewise local-linear models provide a natural way to extend local-linear models to explain the global behavior of the model.

Disease-Atlas: Navigating Disease Trajectories with Deep Learning

no code implementations27 Mar 2018 Bryan Lim, Mihaela van der Schaar

Joint models for longitudinal and time-to-event data are commonly used in longitudinal studies to forecast disease trajectories over time.

AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning

no code implementations ICML 2018 Ahmed M. Alaa, Mihaela van der Schaar

AUTOPROGNOSIS optimizes ensembles of pipeline configurations efficiently using a novel batched Bayesian optimization (BO) algorithm that learns a low-dimensional decomposition of the pipelines high-dimensional hyperparameter space in concurrence with the BO procedure.

Meta-Learning

RadialGAN: Leveraging multiple datasets to improve target-specific predictive models using Generative Adversarial Networks

1 code implementation ICML 2018 Jinsung Yoon, James Jordon, Mihaela van der Schaar

Training complex machine learning models for prediction often requires a large amount of data that is not always readily available.

GANITE: Estimation of Individualized Treatment Effects using Generative Adversarial Nets

1 code implementation ICLR 2018 Jinsung Yoon, James Jordon, Mihaela van der Schaar

Estimating individualized treatment effects (ITE) is a challenging task due to the need for an individual's potential outcomes to be learned from biased data and without having access to the counterfactuals.

Deep Sensing: Active Sensing using Multi-directional Recurrent Neural Networks

1 code implementation ICLR 2018 Jinsung Yoon, William R. Zame, Mihaela van der Schaar

At runtime, the operator prescribes a performance level or a cost constraint, and Deep Sensing determines what measurements to take and what to infer from those measurements, and then issues predictions.

Bayesian Nonparametric Causal Inference: Information Rates and Learning Algorithms

no code implementations24 Dec 2017 Ahmed M. Alaa, Mihaela van der Schaar

We investigate the problem of estimating the causal effect of a treatment on individual subjects from observational data, this is a central problem in various application domains, including healthcare, social sciences, and online advertising.

Causal Inference Selection bias

DPSCREEN: Dynamic Personalized Screening

no code implementations NeurIPS 2017 Kartik Ahuja, William Zame, Mihaela van der Schaar

However, there has been limited work to address the personalized screening for these different diseases.

Estimating Missing Data in Temporal Data Streams Using Multi-directional Recurrent Neural Networks

1 code implementation23 Nov 2017 Jinsung Yoon, William R. Zame, Mihaela van der Schaar

Existing methods address this estimation problem by interpolating within data streams or imputing across data streams (both of which ignore important information) or ignoring the temporal aspect of the data and imposing strong assumptions about the nature of the data-generating process and/or the pattern of missing data (both of which are especially problematic for medical data).

Matrix Completion Multivariate Time Series Imputation

Deep Counterfactual Networks with Propensity-Dropout

1 code implementation19 Jun 2017 Ahmed M. Alaa, Michael Weisz, Mihaela van der Schaar

The network is trained in alternating phases, where in each phase we use the training examples of one of the two potential outcomes (treated and control populations) to update the weights of the shared layers and the respective outcome-specific layers.

Causal Inference Selection bias

ToPs: Ensemble Learning with Trees of Predictors

no code implementations5 Jun 2017 Jinsung Yoon, William R. Zame, Mihaela van der Schaar

Our approach constructs a tree of subsets of the feature space and associates a predictor (predictive model) - determined by training one of a given family of base learners on an endogenously determined training set - to each node of the tree; we call the resulting object a tree of predictors.

Ensemble Learning

Individualized Risk Prognosis for Critical Care Patients: A Multi-task Gaussian Process Model

no code implementations22 May 2017 Ahmed M. Alaa, Jinsung Yoon, Scott Hu, Mihaela van der Schaar

We report the development and validation of a data-driven real-time risk score that provides timely assessments for the clinical acuity of ward patients based on their temporal lab tests and vital signs, which allows for timely intensive care unit (ICU) admissions.

Learning from Clinical Judgments: Semi-Markov-Modulated Marked Hawkes Processes for Risk Prognosis

no code implementations ICML 2017 Ahmed M. Alaa, Scott Hu, Mihaela van der Schaar

Our model captures "informatively sampled" patient episodes: the clinicians' decisions on when to observe a hospitalized patient's vital signs and lab tests over time are represented by a marked Hawkes process, with intensity parameters that are modulated by the patient's latent clinical states, and with observable physiological data (mark process) modeled as a switching multi-task Gaussian process.

Context-Aware Hierarchical Online Learning for Performance Maximization in Mobile Crowdsourcing

no code implementations10 May 2017 Sabrina Klos, Cem Tekin, Mihaela van der Schaar, Anja Klein

In our algorithm, a local controller (LC) in the mobile device of a worker regularly observes the worker's context, her/his decisions to accept or decline tasks and the quality in completing tasks.

Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes

1 code implementation NeurIPS 2017 Ahmed M. Alaa, Mihaela van der Schaar

Stemming from the potential outcomes model, we propose a novel multi- task learning framework in which factual and counterfactual outcomes are mod- eled as the outputs of a function in a vector-valued reproducing kernel Hilbert space (vvRKHS).

Bayesian Inference Gaussian Processes +2

Constructing Effective Personalized Policies Using Counterfactual Inference from Biased Data Sets with Many Features

no code implementations23 Dec 2016 Onur Atan, William R. Zame, Qiaojun Feng, Mihaela van der Schaar

This paper proposes a novel approach for constructing effective personalized policies when the observed data lacks counter-factual information, is biased and possesses many features.

Counterfactual Inference

A Hidden Absorbing Semi-Markov Model for Informatively Censored Temporal Data: Learning and Inference

no code implementations18 Dec 2016 Ahmed M. Alaa, Mihaela van der Schaar

Modeling continuous-time physiological processes that manifest a patient's evolving clinical states is a key step in approaching many problems in healthcare.

A Non-parametric Learning Method for Confidently Estimating Patient's Clinical State and Dynamics

no code implementations NeurIPS 2016 William Hoiles, Mihaela van der Schaar

Estimating patient's clinical state from multiple concurrent physiological streams plays an important role in determining if a therapeutic intervention is necessary and for triaging patients in the hospital.

Bayesian Inference

A Semi-Markov Switching Linear Gaussian Model for Censored Physiological Data

no code implementations16 Nov 2016 Ahmed M. Alaa, Jinsung Yoon, Scott Hu, Mihaela van der Schaar

Critically ill patients in regular wards are vulnerable to unanticipated clinical dete- rioration which requires timely transfer to the intensive care unit (ICU).

Personalized Donor-Recipient Matching for Organ Transplantation

no code implementations12 Nov 2016 Jinsung Yoon, Ahmed M. Alaa, Martin Cadeiras, Mihaela van der Schaar

Organ transplants can improve the life expectancy and quality of life for the recipient but carries the risk of serious post-operative complications, such as septic shock and organ rejection.

Personalized Risk Scoring for Critical Care Prognosis using Mixtures of Gaussian Processes

no code implementations27 Oct 2016 Ahmed M. Alaa, Jinsung Yoon, Scott Hu, Mihaela van der Schaar

Objective: In this paper, we develop a personalized real-time risk scoring algorithm that provides timely and granular assessments for the clinical acuity of ward patients based on their (temporal) lab tests and vital signs; the proposed risk scoring system ensures timely intensive care unit (ICU) admissions for clinically deteriorating patients.

Gaussian Processes Transfer Learning

Context-Aware Proactive Content Caching with Service Differentiation in Wireless Networks

no code implementations14 Jun 2016 Sabrina Müller, Onur Atan, Mihaela van der Schaar, Anja Klein

We derive a sublinear regret bound, which characterizes the learning speed and proves that our algorithm converges to the optimal cache content placement strategy in terms of maximizing the number of cache hits.

Personalized Risk Scoring for Critical Care Patients using Mixtures of Gaussian Process Experts

no code implementations3 May 2016 Ahmed M. Alaa, Jinsung Yoon, Scott Hu, Mihaela van der Schaar

We develop a personalized real time risk scoring algorithm that provides timely and granular assessments for the clinical acuity of ward patients based on their (temporal) lab tests and vital signs.

Transfer Learning

Data-Driven Online Decision Making with Costly Information Acquisition

no code implementations11 Feb 2016 Onur Atan, Mihaela van der Schaar

In most real-world settings such as recommender systems, finance, and healthcare, collecting useful information is costly and requires an active choice on the part of the decision maker.

Decision Making Recommendation Systems

ConfidentCare: A Clinical Decision Support System for Personalized Breast Cancer Screening

no code implementations1 Feb 2016 Ahmed M. Alaa, Kyeong H. Moon, William Hsu, Mihaela van der Schaar

A cluster of patients is a set of patients with similar features (e. g. age, breast density, family history, etc.

Personalized Course Sequence Recommendations

no code implementations30 Dec 2015 Jie Xu, Tianwei Xing, Mihaela van der Schaar

Given the variability in student learning it is becoming increasingly important to tailor courses as well as course sequences to student needs.

Multi-Armed Bandits

Adaptive Ensemble Learning with Confidence Bounds

no code implementations23 Dec 2015 Cem Tekin, Jinsung Yoon, Mihaela van der Schaar

Extracting actionable intelligence from distributed, heterogeneous, correlated and high-dimensional data sources requires run-time processing and learning both locally and globally.

Ensemble Learning Meta-Learning

Predicting Grades

no code implementations16 Aug 2015 Yannick Meier, Jie Xu, Onur Atan, Mihaela van der Schaar

We derive a confidence estimate for the prediction accuracy and demonstrate the performance of our algorithm on a dataset obtained based on the performance of approximately 700 UCLA undergraduate students who have taken an introductory digital signal processing over the past 7 years.

Episodic Multi-armed Bandits

no code implementations4 Aug 2015 Cem Tekin, Mihaela van der Schaar

After the $stop$ action is taken, the learner collects a terminal reward, and observes the costs and terminal rewards associated with each step of the episode.

Multi-Armed Bandits

Global Bandits

no code implementations29 Mar 2015 Onur Atan, Cem Tekin, Mihaela van der Schaar

In the case in which rewards of all arms are deterministic functions of a single unknown parameter, we construct a greedy policy that achieves {\em bounded regret}, with a bound that depends on the single true parameter of the problem.

Decision Making Informativeness +1

Contextual Online Learning for Multimedia Content Aggregation

no code implementations7 Feb 2015 Cem Tekin, Mihaela van der Schaar

A key challenge for such systems is to accurately predict what type of content each of its consumers prefers in a certain context, and adapt these predictions to the evolving consumers' preferences, contexts and content characteristics.

RELEAF: An Algorithm for Learning and Exploiting Relevance

no code implementations5 Feb 2015 Cem Tekin, Mihaela van der Schaar

We prove a general regret bound for our algorithm whose time order depends only on the maximum number of relevant dimensions among all the actions, which in the special case where the relevance relation is single-valued (a function), reduces to $\tilde{O}(T^{2(\sqrt{2}-1)})$; in the absence of a relevance relation, the best known contextual bandit algorithms achieve regret $\tilde{O}(T^{(D+1)/(D+2)})$, where $D$ is the full dimension of the context vector.

Decision Making Medical Diagnosis +1

Discovering, Learning and Exploiting Relevance

no code implementations NeurIPS 2014 Cem Tekin, Mihaela van der Schaar

When the relation is a function, i. e., the reward of an action only depends on the context of a single type, and the expected reward of an action is Lipschitz continuous in the context of its relevant type, we propose an algorithm that achieves $\tilde{O}(T^{\gamma})$ regret with a high probability, where $\gamma=2/(1+\sqrt{2})$.

Medical Diagnosis Recommendation Systems

Jamming Bandits

no code implementations13 Nov 2014 SaiDhiraj Amuru, Cem Tekin, Mihaela van der Schaar, R. Michael Buehrer

We first present novel online learning algorithms to maximize the jamming efficacy against static transmitter-receiver pairs and prove that our learning algorithm converges to the optimal (in terms of the error rate inflicted at the victim and the energy used) jamming strategy.

Global Bandits with Holder Continuity

no code implementations29 Oct 2014 Onur Atan, Cem Tekin, Mihaela van der Schaar

Specifically, we prove that the parameter-free (worst-case) regret is sublinear in time, and decreases with the informativeness of the arms.

Informativeness

Forecasting Popularity of Videos using Social Media

no code implementations22 Mar 2014 Jie Xu, Mihaela van der Schaar, Jiangchuan Liu, Haitao Li

This paper presents a systematic online prediction method (Social-Forecast) that is capable to accurately forecast the popularity of videos promoted by social media.

Distributed Online Learning in Social Recommender Systems

no code implementations26 Sep 2013 Cem Tekin, Simpson Zhang, Mihaela van der Schaar

In contrast to centralized recommender systems, in which there is a single centralized seller who has access to the complete inventory of items as well as the complete record of sales and user information, in decentralized recommender systems each seller/learner only has access to the inventory of items and user information for its own products and not the products and user information of other sellers, but can get commission if it sells an item of another seller.

Decision Making Recommendation Systems

Ensemble of Distributed Learners for Online Classification of Dynamic Data Streams

no code implementations24 Aug 2013 Luca Canzian, Yu Zhang, Mihaela van der Schaar

We present an efficient distributed online learning scheme to classify data captured from distributed, heterogeneous, and dynamic data sources.

Ensemble Learning General Classification

Distributed Online Learning via Cooperative Contextual Bandits

no code implementations21 Aug 2013 Cem Tekin, Mihaela van der Schaar

At each moment of time, an instance characterized by a certain context may arrive to each learner; based on the context, the learner can select one of its own actions (which gives a reward and provides information) or request assistance from another learner.

Event Detection Multi-Armed Bandits +1

Decentralized Online Big Data Classification - a Bandit Framework

no code implementations21 Aug 2013 Cem Tekin, Mihaela van der Schaar

We model the problem of joint classification by the distributed and heterogeneous learners from multiple data sources as a distributed contextual bandit problem where each data is characterized by a specific context.

General Classification

Distributed Online Big Data Classification Using Context Information

no code implementations2 Jul 2013 Cem Tekin, Mihaela van der Schaar

Distributed, online data mining systems have emerged as a result of applications requiring analysis of large amounts of correlated and high-dimensional data produced by multiple distributed data sources.

General Classification

Fast Reinforcement Learning for Energy-Efficient Wireless Communications

no code implementations29 Sep 2010 Nicholas Mastronarde, Mihaela van der Schaar

The advantages of the proposed online method are that (i) it does not require a priori knowledge of the traffic arrival and channel statistics to determine the jointly optimal power-control, AMC, and DPM policies; (ii) it exploits partial information about the system so that less information needs to be learned than when using conventional reinforcement learning algorithms; and (iii) it obviates the need for action exploration, which severely limits the adaptation speed and run-time performance of conventional reinforcement learning algorithms.

Stochastic Optimization

Cannot find the paper you are looking for? You can Submit a new open access paper.