Publications

Modality Translation for Object Detection Adaptation Without Forgetting Prior Knowledge
Heitor Rapela Medeiros
Masih Aminbeidokhti
Fidel A. Guerrero Peña
David Latortue
Eric Granger
Marco Pedersoli
A common practice in deep learning involves training large neural networks on massive datasets to achieve high accuracy across various domai… (voir plus)ns and tasks. While this approach works well in many application areas, it often fails drastically when processing data from a new modality with a significant distribution shift from the data used to pre-train the model. This paper focuses on adapting a large object detection model trained on RGB images to new data extracted from IR images with a substantial modality shift. We propose Modality Translator (ModTr) as an alternative to the common approach of fine-tuning a large model to the new modality. ModTr adapts the IR input image with a small transformation network trained to directly minimize the detection loss. The original RGB model can then work on the translated inputs without any further changes or fine-tuning to its parameters. Experimental results on translating from IR to RGB images on two well-known datasets show that our simple approach provides detectors that perform comparably or better than standard fine-tuning, without forgetting the knowledge of the original model. This opens the door to a more flexible and efficient service-based detection pipeline, where a unique and unaltered server, such as an RGB detector, runs constantly while being queried by different modalities, such as IR with the corresponding translations model. Our code is available at: https://github.com/heitorrapela/ModTr.
2024-11-23
Lecture Notes in Computer Science (publié)
doi.org
arxiv.org
Gaps Between Research and Practice When Measuring Representational Harms Caused by LLM-Based Systems
Emma Harvey
Emily Sheng
Su Lin Blodgett
Alexandra Chouldechova
Jean Garcia-Gathright
A.R. Olteanu
Hanna Wallach
To facilitate the measurement of representational harms caused by large language model (LLM)-based systems, the NLP research community has p… (voir plus)roduced and made publicly available numerous measurement instruments, including tools, datasets, metrics, benchmarks, annotation instructions, and other techniques. However, the research community lacks clarity about whether and to what extent these instruments meet the needs of practitioners tasked with developing and deploying LLM-based systems in the real world, and how these instruments could be improved. Via a series of semi-structured interviews with practitioners in a variety of roles in different organizations, we identify four types of challenges that prevent practitioners from effectively using publicly available instruments for measuring representational harms caused by LLM-based systems: (1) challenges related to using publicly available measurement instruments; (2) challenges related to doing measurement in practice; (3) challenges arising from measurement tasks involving LLM-based systems; and (4) challenges specific to measuring representational harms. Our goal is to advance the development of instruments for measuring representational harms that are well-suited to practitioner needs, thus better facilitating the responsible development and deployment of LLM-based systems.
2024-11-22
ArXiv (prépublication)
doi.org
arxiv.org
Learning and fine-tuning a generic value-selection heuristic inside a constraint programming solver
Tom Marty
Léo Boisvert
Tristan François
Pierre Tessier
Louis Gautier
Louis-Martin Rousseau
Quentin Cappart
Constraint programming is known for being an efficient approach to solving combinatorial problems. Important design choices in a solver are … (voir plus)the branching heuristics, designed to lead the search to the best solutions in a minimum amount of time. However, developing these heuristics is a time-consuming process that requires problem-specific expertise. This observation has motivated many efforts to use machine learning to automatically learn efficient heuristics without expert intervention. Although several generic variable-selection heuristics are available in the literature, the options for value-selection heuristics are more scarce. We propose to tackle this issue by introducing a generic learning procedure that can be used to obtain a value-selection heuristic inside a constraint programming solver. This has been achieved thanks to the combination of a deep Q-learning algorithm, a tailored reward signal, and a heterogeneous graph neural network. Experiments on graph coloring, maximum independent set, maximum cut, and minimum vertex cover problems show that this framework competes with the well-known impact-based and activity-based search heuristics and can find solutions close to optimality without requiring a large number of backtracks. Additionally, we observe that fine-tuning a model with a different problem class can accelerate the learning process.
2024-11-22
Constraints (publié)
doi.org
MixEHR-Nest: Identifying Subphenotypes within Electronic Health Records through Hierarchical Guided-Topic Modeling
Ruohan Wang
Zilong Wang
Ziyang Song
David Buckeridge
Yue Li
Automatic subphenotyping from electronic health records (EHRs)provides numerous opportunities to understand diseases with unique subgroups a… (voir plus)nd enhance personalized medicine for patients. However, existing machine learning algorithms either focus on specific diseases for better interpretability or produce coarse-grained phenotype topics without considering nuanced disease patterns. In this study, we propose a guided topic model, MixEHR-Nest, to infer sub-phenotype topics from thousands of disease using multi-modal EHR data. Specifically, MixEHR-Nest detects multiple subtopics from each phenotype topic, whose prior is guided by the expert-curated phenotype concepts such as Phenotype Codes (PheCodes) or Clinical Classification Software (CCS) codes. We evaluated MixEHR-Nest on two EHR datasets: (1) the MIMIC-III dataset consisting of over 38 thousand patients from intensive care unit (ICU) from Beth Israel Deaconess Medical Center (BIDMC) in Boston, USA; (2) the healthcare administrative database PopHR, comprising 1.3 million patients from Montreal, Canada. Experimental results demonstrate that MixEHR-Nest can identify subphenotypes with distinct patterns within each phenotype, which are predictive for disease progression and severity. Consequently, MixEHR-Nest distinguishes between type 1 and type 2 diabetes by inferring subphenotypes using CCS codes, which do not differentiate these two subtype concepts. Additionally, MixEHR-Nest not only improved the prediction accuracy of short-term mortality of ICU patients and initial insulin treatment in diabetic patients but also revealed the contributions of subphenotypes. For longitudinal analysis, MixEHR-Nest identified subphenotypes of distinct age prevalence under the same phenotypes, such as asthma, leukemia, epilepsy, and depression. The MixEHR-Nest software is available at GitHub: https://github.com/li-lab-mcgill/MixEHR-Nest.
2024-11-21
Proceedings of the 15th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics (publié)
doi.org
arxiv.org
scMoE: single-cell mixture of experts for learning hierarchical, cell-type-specific, and interpretable representations from heterogeneous scRNA-seq data
Michael Huang
Yue Li
Advancements in single-cell transcriptomics methods have resulted in a wealth of single-cell RNA sequencing (scRNA-seq) data. Methods to lea… (voir plus)rn cell representation from atlas-level scRNA-seq data across diverse tissues can shed light into cell functions implicated in diseases such as cancer. However, integrating large-scale and heterogeneous scRNA-seq data is challenging due to the disparity of cell-types and batch effects. We present single-cell Mixture of Expert (scMoE), a hierarchical mixture of experts single-cell topic model. Our key contributions are the cell-type specific experts, which explicitly aligns topics with cell-types, and the integration of hierarchical cell-type lineages and domain knowledge. scMoE is both transferable and highly interpretable. We benchmarked our scMoE’s performance on 9 single-cell RNA-seq datasets for clustering and 3 simulated spatial datasets for spatial deconvolution. We additionally show that our model, using single-cell references, yields meaningful biological results by deconvolving 3 cancer bulk RNA-seq datasets and 2 spatial transcriptomics datasets. scMoE is able to identify cell-types of survival importance, find cancer subtype specific deconvolutional patterns, and capture meaningful spatially distinct cell-type distributions.
2024-11-21
Proceedings of the 15th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics (publié)
doi.org
Improving Geo-diversity of Generated Images with Contextualized Vendi Score Guidance
Reyhane Askari Hemmat
Melissa Hall
Alicia Sun
Candace Ross
Michal Drozdzal
Adriana Romero
2024-11-20
Lecture Notes in Computer Science (publié)
doi.org
arxiv.org
Effectiveness of primary repair for low anorectal malformations in Uganda.
Felix Oyania
Sarah Ullrich
Zane Hellmann
Caroline Q. Stephens
Meera Kotagal
Sarah Jane Commander
Amy M. Shui
Martin Situma
Charles Newton Odongo
Olivia Kituuka
Francis Bajunirwe
Doruk Ozgediz
Dan Poenaru
2024-11-18
Pediatric surgery international (Print) (publié)
doi.org
Estimation and group-feature selection in sparse mixture-of-experts with diverging number of parameters
Abbas Khalili
Archer Yi Yang
Da Xu
2024-11-18
Journal of Statistical Planning and Inference (publié)
doi.org
Exploring the Manifold of Neural Networks Using Diffusion Geometry
Elliott Abel
Peyton Crevasse
Yvan Grinspan
Selma Mazioud
Folu Ogundipe
Kristof Reimann
Ellie Schueler
Andrew J. Steindl
Ellen Zhang
Dhananjay Bhaskar
Siddharth Viswanath
Yanlei Zhang
Tim G. J. Rudner
Ian Adelstein
Smita Krishnaswamy
Drawing motivation from the manifold hypothesis, which posits that most high-dimensional data lies on or near low-dimensional manifolds, we … (voir plus)apply manifold learning to the space of neural networks. We learn manifolds where datapoints are neural networks by introducing a distance between the hidden layer representations of the neural networks. These distances are then fed to the non-linear dimensionality reduction algorithm PHATE to create a manifold of neural networks. We characterize this manifold using features of the representation, including class separation, hierarchical cluster structure, spectral entropy, and topological structure. Our analysis reveals that high-performing networks cluster together in the manifold, displaying consistent embedding patterns across all these features. Finally, we demonstrate the utility of this approach for guiding hyperparameter optimization and neural architecture search by sampling from the manifold.
2024-11-18
ArXiv (prépublication)
doi.org
arxiv.org
Prediction of Post Traumatic Epilepsy Using <scp>MR</scp>‐Based Imaging Markers
Haleh Akrami
Wenhui Cui
Paul E. Kim
Christianne Heck
Andrei Irimia
Karim Jerbi
Dileep Nair
Richard M. Leahy
Anand A. Joshi
Post-traumatic epilepsy (PTE) is a debilitating neurological disorder that develops after traumatic brain injury (TBI). Despite the high pre… (voir plus)valence of PTE, current methods for predicting its occurrence remain limited. In this study, we aimed to identify imaging-based markers for the prediction of PTE using machine learning. Specifically, we examined three imaging features: Lesion volumes, resting-state fMRI-based measures of functional connectivity, and amplitude of low-frequency fluctuation (ALFF). We employed three machine-learning methods, namely, kernel support vector machine (KSVM), random forest, and an artificial neural network (NN), to develop predictive models. Our results showed that the KSVM classifier, with all three feature types as input, achieved the best prediction accuracy of 0.78 AUC (area under the receiver operating characteristic (ROC) curve) using nested cross-validation. Furthermore, we performed voxel-wise and lobe-wise group difference analyses to investigate the specific brain regions and features that the model found to be most helpful in distinguishing PTE from non-PTE populations. Our statistical analysis uncovered significant differences in bilateral temporal lobes and cerebellum between PTE and non-PTE groups. Overall, our findings demonstrate the complementary prognostic value of MR-based markers in PTE prediction and provide new insights into the underlying structural and functional alterations associated with PTE.
2024-11-18
Human Brain Mapping (publié)
doi.org
Sketch-guided Cage-based 3D Gaussian Splatting Deformation
Tianhao Xie
Noam Aigerman
Eugene Belilovsky
Tiberiu Popa
3D Gaussian Splatting (GS) is one of the most promising novel 3D representations that has received great interest in computer graphics and c… (voir plus)omputer vision. While various systems have introduced editing capabilities for 3D GS, such as those guided by text prompts, fine-grained control over deformation remains an open challenge. In this work, we present a novel sketch-guided 3D GS deformation system that allows users to intuitively modify the geometry of a 3D GS model by drawing a silhouette sketch from a single viewpoint. Our approach introduces a new deformation method that combines cage-based deformations with a variant of Neural Jacobian Fields, enabling precise, fine-grained control. Additionally, it leverages large-scale 2D diffusion priors and ControlNet to ensure the generated deformations are semantically plausible. Through a series of experiments, we demonstrate the effectiveness of our method and showcase its ability to animate static 3D GS models as one of its key applications.
2024-11-18
ArXiv (prépublication)
doi.org
arxiv.org
EDAI Framework for Integrating Equity, Diversity, and Inclusion Throughout the Lifecycle of AI to Improve Health and Oral Health Care: Qualitative Study
Samira Abbasgholizadeh Rahimi
Richa Shrivastava
Anita Brown-Johnson
Pascale Caidor
Claire Davies
Amal Idrissi Janati
Pascaline Kengne Talla
Sreenath Madathil
Bettina M Willie
Elham Emami
Recent studies have identified significant gaps in equity, diversity, and inclusion (EDI) considerations within the lifecycle of artificial … (voir plus)intelligence (AI), spanning from data collection and problem definition to implementation stages. Despite the recognized need for integrating EDI principles, there is currently no existing guideline or framework to support this integration in the AI lifecycle. This study aimed to address this gap by identifying EDI principles and indicators to be integrated into the AI lifecycle. The goal was to develop a comprehensive guiding framework to guide the development and implementation of future AI systems. This study was conducted in 3 phases. In phase 1, a comprehensive systematic scoping review explored how EDI principles have been integrated into AI in health and oral health care settings. In phase 2, a multidisciplinary team was established, and two 2-day, in-person international workshops with over 60 representatives from diverse backgrounds, expertise, and communities were conducted. The workshops included plenary presentations, round table discussions, and focused group discussions. In phase 3, based on the workshops’ insights, the EDAI framework was developed and refined through iterative feedback from participants. The results of the initial systematic scoping review have been published separately, and this paper focuses on subsequent phases of the project, which is related to framework development. In this study, we developed the EDAI framework, a comprehensive guideline that integrates EDI principles and indicators throughout the entire AI lifecycle. This framework addresses existing gaps at various stages, from data collection to implementation, and focuses on individual, organizational, and systemic levels. Additionally, we identified both the facilitators and barriers to integrating EDI within the AI lifecycle in health and oral health care. The developed EDAI framework provides a comprehensive, actionable guideline for integrating EDI principles into AI development and deployment. By facilitating the systematic incorporation of these principles, the framework supports the creation and implementation of AI systems that are not only technologically advanced but also sensitive to EDI principles.
2024-11-14
Journal of Medical Internet Research (publié)
doi.org