Benjamin M. Marlin
YOU?
Author Swipe
View article: A dynamic Bayesian network approach to modeling engagement and walking behavior: insights from a yearlong micro-randomized trial ( <i>Heartsteps II</i> )
A dynamic Bayesian network approach to modeling engagement and walking behavior: insights from a yearlong micro-randomized trial ( <i>Heartsteps II</i> ) Open
The results demonstrate the benefits of DBNs to model the daily-level idiographic behavioral dynamics of engagement in digital intervention studies. This approach can be leveraged to support the refinement of dynamic theories of behavior c…
View article: ACE and Diverse Generalization via Selective Disagreement
ACE and Diverse Generalization via Selective Disagreement Open
Deep neural networks are notoriously sensitive to spurious correlations - where a model learns a shortcut that fails out-of-distribution. Existing work on spurious correlations has often focused on incomplete correlations,leveraging access…
View article: SigmaScheduling: Uncertainty-Informed Scheduling of Decision Points for Intelligent Mobile Health Interventions
SigmaScheduling: Uncertainty-Informed Scheduling of Decision Points for Intelligent Mobile Health Interventions Open
Timely decision making is critical to the effectiveness of mobile health (mHealth) interventions. At predefined timepoints called "decision points," intelligent mHealth systems such as just-in-time adaptive interventions (JITAIs) estimate …
View article: Enhancing Adaptive Behavioral Interventions with LLM Inference from Participant-Described States
Enhancing Adaptive Behavioral Interventions with LLM Inference from Participant-Described States Open
The use of reinforcement learning (RL) methods to support health behavior change via personalized and just-in-time adaptive interventions is of significant interest to health and behavioral science researchers focused on problems such as s…
View article: BOTS: Batch Bayesian Optimization of Extended Thompson Sampling for Severely Episode-Limited RL Settings
BOTS: Batch Bayesian Optimization of Extended Thompson Sampling for Severely Episode-Limited RL Settings Open
In settings where the application of reinforcement learning (RL) requires running real-world trials, including the optimization of adaptive health interventions, the number of episodes available for learning can be severely limited due to …
View article: StepCountJITAI: simulation environment for RL with application to physical activity adaptive intervention
StepCountJITAI: simulation environment for RL with application to physical activity adaptive intervention Open
The use of reinforcement learning (RL) to learn policies for just-in-time adaptive interventions (JITAIs) is of significant interest in many behavioral intervention domains including improving levels of physical activity. In a messaging-ba…
View article: Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation
Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation Open
Wearable sensors enable health researchers to continuously collect data pertaining to the physiological state of individuals in real-world settings. However, such data can be subject to extensive missingness due to a complex combination of…
View article: FlexLoc: Conditional Neural Networks for Zero-Shot Sensor Perspective Invariance in Object Localization with Distributed Multimodal Sensors
FlexLoc: Conditional Neural Networks for Zero-Shot Sensor Perspective Invariance in Object Localization with Distributed Multimodal Sensors Open
Localization is a critical technology for various applications ranging from navigation and surveillance to assisted living. Localization systems typically fuse information from sensors viewing the scene from different perspectives to estim…
View article: GDTM: An Indoor Geospatial Tracking Dataset with Distributed Multimodal Sensors
GDTM: An Indoor Geospatial Tracking Dataset with Distributed Multimodal Sensors Open
Constantly locating moving objects, i.e., geospatial tracking, is essential for autonomous building infrastructure. Accurate and robust geospatial tracking often leverages multimodal sensor fusion algorithms, which require large datasets w…
View article: Artificial Intelligence and Technology Collaboratories: Innovating aging research and Alzheimer's care
Artificial Intelligence and Technology Collaboratories: Innovating aging research and Alzheimer's care Open
This perspective outlines the Artificial Intelligence and Technology Collaboratories (AITC) at Johns Hopkins University, University of Pennsylvania, and University of Massachusetts, highlighting their roles in developing AI‐based technolog…
View article: REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learning
REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learning Open
The success of self-supervised contrastive learning hinges on identifying positive data pairs, such that when they are pushed together in embedding space, the space encodes useful information for subsequent downstream tasks. Constructing p…
View article: Heteroskedastic Geospatial Tracking with Distributed Camera Networks
Heteroskedastic Geospatial Tracking with Distributed Camera Networks Open
Visual object tracking has seen significant progress in recent years. However, the vast majority of this work focuses on tracking objects within the image plane of a single camera and ignores the uncertainty associated with predicted objec…
View article: Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions
Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions Open
Just-in-Time Adaptive Interventions (JITAIs) are a class of personalized health interventions developed within the behavioral science community. JITAIs aim to provide the right type and amount of support by iteratively selecting a sequence…
View article: The ILHBN: challenges, opportunities, and solutions from harmonizing data under heterogeneous study designs, target populations, and measurement protocols
The ILHBN: challenges, opportunities, and solutions from harmonizing data under heterogeneous study designs, target populations, and measurement protocols Open
The ILHBN is funded by the National Institutes of Health to collaboratively study the interactive dynamics of behavior, health, and the environment using Intensive Longitudinal Data (ILD) to (a) understand and intervene on behavior and hea…
View article: Design and Deployment of a Multi-Modal Multi-Node Sensor Data Collection Platform
Design and Deployment of a Multi-Modal Multi-Node Sensor Data Collection Platform Open
Sensing and data collection platforms are the crucial components of high-quality datasets that can fuel advancements in research. However, such platforms usually are ad-hoc designs and are limited in sensor modalities. In this paper, we di…
View article: BayesLDM: A Domain-specific Modeling Language for Probabilistic Modeling of Longitudinal Data.
BayesLDM: A Domain-specific Modeling Language for Probabilistic Modeling of Longitudinal Data. Open
In this paper we present BayesLDM, a library for Bayesian longitudinal data modeling consisting of a high-level modeling language with specific features for modeling complex multivariate time series data coupled with a compiler that can pr…
View article: BayesLDM: A Domain-Specific Language for Probabilistic Modeling of Longitudinal Data
BayesLDM: A Domain-Specific Language for Probabilistic Modeling of Longitudinal Data Open
In this paper we present BayesLDM, a system for Bayesian longitudinal data modeling consisting of a high-level modeling language with specific features for modeling complex multivariate time series data coupled with a compiler that can pro…
View article: Advancing Behavioral Intervention and Theory Development for Mobile Health: The HeartSteps II Protocol
Advancing Behavioral Intervention and Theory Development for Mobile Health: The HeartSteps II Protocol Open
Background: Recent advances in mobile and wearable technologies have led to new forms of interventions, called “Just-in-Time Adaptive Interventions” (JITAI). JITAIs interact with the individual at the most appropriate time and provide the …
View article: Impact of Parameter Sparsity on Stochastic Gradient MCMC Methods for Bayesian Deep Learning
Impact of Parameter Sparsity on Stochastic Gradient MCMC Methods for Bayesian Deep Learning Open
Bayesian methods hold significant promise for improving the uncertainty quantification ability and robustness of deep neural network models. Recent research has seen the investigation of a number of approximate Bayesian inference methods f…
View article: Challenges and Opportunities in Approximate Bayesian Deep Learning for Intelligent IoT Systems
Challenges and Opportunities in Approximate Bayesian Deep Learning for Intelligent IoT Systems Open
Approximate Bayesian deep learning methods hold significant promise for addressing several issues that occur when deploying deep learning components in intelligent systems, including mitigating the occurrence of over-confident errors and p…
View article: Post-hoc loss-calibration for Bayesian neural networks
Post-hoc loss-calibration for Bayesian neural networks Open
Bayesian decision theory provides an elegant framework for acting optimally under uncertainty when tractable posterior distributions are available. Modern Bayesian models, however, typically involve intractable posteriors that are approxim…
View article: Heteroscedastic Temporal Variational Autoencoder For Irregularly Sampled Time Series.
Heteroscedastic Temporal Variational Autoencoder For Irregularly Sampled Time Series. Open
Irregularly sampled time series commonly occur in several domains where they present a significant challenge to standard deep learning models. In this paper, we propose a new deep learning framework for probabilistic interpolation of irreg…
View article: Heteroscedastic Temporal Variational Autoencoder For Irregular Time Series
Heteroscedastic Temporal Variational Autoencoder For Irregular Time Series Open
Irregularly sampled time series commonly occur in several domains where they present a significant challenge to standard deep learning models. In this paper, we propose a new deep learning framework for probabilistic interpolation of irreg…
View article: Post-hoc loss-calibration for Bayesian neural networks
Post-hoc loss-calibration for Bayesian neural networks Open
Bayesian decision theory provides an elegant framework for acting optimally under uncertainty when tractable posterior distributions are available. Modern Bayesian models, however, typically involve intractable posteriors that are approxim…
View article: Multi-Time Attention Networks for Irregularly Sampled Time Series
Multi-Time Attention Networks for Irregularly Sampled Time Series Open
Irregular sampling occurs in many time series modeling applications where it presents a significant challenge to standard deep learning models. This work is motivated by the analysis of physiological time series data in electronic health r…