More Like this

1. Gaussian graphical model‐based heterogeneity analysis via penalized fusion

   https://doi.org/10.1111/biom.13426  

   Ren, Mingyang; Zhang, Sanguo; Zhang, Qingzhao; Ma, Shuangge (January 2021, Biometrics)

   null (Ed.)

   Heterogeneity is a hallmark of cancer, diabetes, cardiovascular diseases, and many other complex diseases. This study has been partly motivated by the unsupervised heterogeneity analysis for complex diseases based on molecular and imaging data, for which, network-based analysis, by accommodating the interconnections among variables, can be more informative than that limited to mean, variance, and other simple distributional properties. In the literature, there has been very limited research on network-based heterogeneity analysis, and a common limitation shared by the existing techniques is that the number of subgroups needs to be specified a priori or in an ad hoc manner. In this article, we develop a penalized fusion approach for heterogeneity analysis based on the Gaussian graphical model. It applies penalization to the mean and precision matrix parameters to generate regularized and interpretable estimates. More importantly, a fusion penalty is imposed to "automatedly" determine the number of subgroups and generate more concise, reliable, and interpretable estimation. Consistency properties are rigorously established, and an effective computational algorithm is developed. The heterogeneity analysis of non-small-cell lung cancer based on single-cell gene expression data of the Wnt pathway and that of lung adenocarcinoma based on histopathological imaging data not only demonstrate the practical applicability of the proposed approach but also lead to interesting new findings. 

   more » « less
2. Learning a Pedestrian Social Behavior Dictionary

   Johnson, Faith; Dana, K (November 2023, British Machine Vision Conference 2023)

   Understanding pedestrian behavior patterns is key for building autonomous agents that can navigate among humans. We seek a learned dictionary of pedestrian behavior to obtain a semantic description of pedestrian trajectories. Supervised methods for dictionary learning are often impractical since pedestrian behaviors may be unknown a priori and manually generating behavior labels is prohibitively time consuming. We utilize a novel, unsupervised framework to create a taxonomy of pedestrian behavior observed in a specific space. First, we learn a trajectory latent space that enables unsupervised clustering to create an interpretable pedestrian behavior dictionary. Then, we show the utility of this dictionary for building pedestrian behavior maps to visualize space usage patterns and for computing distributions of behaviors in a space. We demonstrate a simple but effective trajectory prediction by conditioning on these behavior labels. While many trajectory analysis methods rely on RNNs or transformers, we develop a lightweight, low-parameter approach and show results outperforming SOTA on the ETH and UCY datasets. 

   more » « less
3. Neural Additive Deep Clustering Survival Machines: An Interpretable Approach for Alzheimer's Disease and Related Dementias

   https://doi.org/10.1002/alz70855_107083  

   Noshin, Kazi; Hou, Bojian; Bolan, Mary Regina; Shen, Li; Zhang, Aidong (December 2025, Alzheimer's & Dementia)

   Abstract BackgroundIn Alzheimer's Disease research, identifying brain regions that influence disease progression is crucial for understanding pathogenesis and developing targeted therapies. Deep learning models have advanced survival analysis in Alzheimer's Disease research, but their “black box” nature limits clinical utility. We developed Neural Additive Deep Clustering Survival Machines (NADCSM), which leverages Neural Additive Models to provide interpretable insights into brain region contributions while maintaining competitive predictive performance. This approach aims to bridge the gap between model performance and clinical utility, potentially accelerating biomarker discovery and therapeutic development. MethodThe genotyping, demographic, and imaging data used in our experiments are sourced from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We utilized ADNI's AV45 Florbetapir PET imaging data to track MCI and early AD progression. This modality is particularly valuable for tracking the progression of mild cognitive impairment (MCI) and early Alzheimer's disease (AD). Our NADCSM framework models survival times using Weibull distributions and enhances interpretability through Neural Additive Models (NAMs)., where each input feature is processed through multiple univariate shape functions parameterized by multilayer perceptions (MLPs). Performance was evaluated using Concordance Index (C Index) for predictive ability and LogRank statistic for survival curve separation and clustering quality. ResultTable 1 compares the performance of DeepCox, DCSM, and NADCSM (ours) on the AV45 dataset using C Index and LogRank metrics. DCSM achieves the highest C Index (0.7789 ± 0.0193), followed by DeepCox (0.7781 ± 0.0306) and NADCSM (0.7772±0.0236), demonstrating strong predictive accuracy. For LogRank, DCSM leads (317.84 ± 31.89), followed by NADCSM (317.84 ± 31.89) and DeepCox (93.4171±9.7990), indicating NADCSM's competitive performance with enhanced interpretability.Figure 1 shows shape functions for the top six brain ROIs identified by NADCSM, including Fusiform (Left), Cerebelum_Crus (Right), and others. The smooth curves capture the relationships between regional amyloid burden and AD progression, highlighting key survival predictors. ConclusionOur NADCSM framework provides an interpretable risk prediction approach, uncovering significant features and explaining their effects on Alzheimer's and Related Dementias (ADRD) progression. By enhancing transparency, this study can advance precision medicine and improve understanding of ADRD‐related challenges. 

   more » « less
4. Bottleneck Problems: An Information and Estimation-Theoretic View

   https://doi.org/10.3390/e22111325  

   Asoodeh, Shahab; Calmon, Flavio P. (November 2020, Entropy)

   Information bottleneck (IB) and privacy funnel (PF) are two closely related optimization problems which have found applications in machine learning, design of privacy algorithms, capacity problems (e.g., Mrs. Gerber’s Lemma), and strong data processing inequalities, among others. In this work, we first investigate the functional properties of IB and PF through a unified theoretical framework. We then connect them to three information-theoretic coding problems, namely hypothesis testing against independence, noisy source coding, and dependence dilution. Leveraging these connections, we prove a new cardinality bound on the auxiliary variable in IB, making its computation more tractable for discrete random variables. In the second part, we introduce a general family of optimization problems, termed “bottleneck problems”, by replacing mutual information in IB and PF with other notions of mutual information, namely f-information and Arimoto’s mutual information. We then argue that, unlike IB and PF, these problems lead to easily interpretable guarantees in a variety of inference tasks with statistical constraints on accuracy and privacy. While the underlying optimization problems are non-convex, we develop a technique to evaluate bottleneck problems in closed form by equivalently expressing them in terms of lower convex or upper concave envelope of certain functions. By applying this technique to a binary case, we derive closed form expressions for several bottleneck problems. 

   more » « less
5. Clustering Data in Secured, Distributed Datasets

   https://doi.org/10.1007/978-3-030-24311-1_40  

   Sayantan Dey, Lee A. (July 2019, The Third International Workshop on Parallel and Distributed Data Mining)

   The massive growth in data generation and collection has brought to the forefront the necessity to develop mechanized methods to analyze and extract information from them. Data clustering is one of the fundamental modes to discover new insights from data. However, high dimensional data has its own challenges where many conventional clustering algorithms fails either in accuracy or scalability. To further complicate the issue, distinct subsets of sensitive data may reside in geographically separated locations with the sensitive nature of the data preventing (or inhibiting) its access for mechanized analysis. Thus, methods to discover information from the collective whole of these secured, distributed data sets that also preserves the integrity of the data must be found. In this paper we develop and assess a distributed algorithm that can cluster geographically separated data while simultaneously preserving the strict privacy requirements of non sharing of protected high dimensional data. We implement our algorithm on the distributed map-reduce based platform Spark and demonstrate its performance by comparing it to the standard data clustering algorithms. 

   more » « less