An official website of the United States government
We consider the problem of predictive modeling from irregularly and sparsely sampled longitudinal data with unknown, complex correlation structures and abrupt discontinuities. To address these challenges, we introduce a novel inducing clusters longitudinal deep kernel Gaussian Process (ICDKGP). ICDKGP approximates the data generating process by a zero-mean GP with a longitudinal deep kernel that models the unknown complex correlation structure in the data and a deterministic non-zero mean function to model the abrupt discontinuities. To improve the scalability and interpretability of ICDKGP, we introduce inducing clusters corresponding to centers of clusters in the training data. We formulate the training of ICDKGP as a constrained optimization problem and derive its evidence lower bound. We introduce a novel relaxation of the resulting problem which under rather mild assumptions yields a solution with error bounded relative to the original problem. We describe the results of extensive experiments demonstrating that ICDKGP substantially outperforms the state-of-the-art longitudinal methods on data with both smoothly and non-smoothly varying outcomes.
more »
« less
Longitudinal Deep Kernel Gaussian Process Regression
Gaussian processes offer an attractive framework for predictive modeling from longitudinal data, i.e., irregularly sampled, sparse observations from a set of individuals over time. However, such methods have two key shortcomings: (i) They rely on ad hoc heuristics or expensive trial and error to choose the effective kernels, and (ii) They fail to handle multilevel correlation structure in the data. We introduce Longitudinal deep kernel Gaussian process regression (L-DKGPR) to overcome these limitations by fully automating the discovery of complex multilevel correlation structure from longitudinal data. Specifically, L-DKGPR eliminates the need for ad hoc heuristics or trial and error using a novel adaptation of deep kernel learning that combines the expressive power of deep neural networks with the flexibility of non-parametric kernel methods. L-DKGPR effectively learns the multilevel correlation with a novel additive kernel that simultaneously accommodates both time-varying and the time-invariant effects. We derive an efficient algorithm to train L-DKGPR using latent space inducing points and variational inference. Results of extensive experiments on several benchmark data sets demonstrate that L-DKGPR significantly outperforms the state-of-the-art longitudinal data analysis (LDA) methods.
more »
« less
- PAR ID:
- 10287280
- Date Published:
- Journal Name:
- Proceedings of the AAAI Conference on Artificial Intelligence
- ISSN:
- 2159-5399
- Page Range / eLocation ID:
- 8556-8564
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Isolating application components is crucial to limit the exposure of sensitive data and code to vulnerabilities in the untrusted components. Process-based isolation is the de facto isolation used in practice, e.g., web browsers. However, it incurs significant performance overhead and is typically infeasible when frequent switches between isolation domains are expected. To address this problem, many intra-process memory isolation techniques have been proposed using novel kernel abstractions, recent CPU extensions (e.g., Intel ® MPK), and software-based fault isolation (e.g., WebAssembly). However, these techniques insufficiently isolate kernel resources, such as file descriptors, or do so by incurring high overheads when resources are accessed. Other work virtualizes the kernel context inside a privileged user space domain, but this is ad-hoc, error-prone, and provides only limited kernel functionalities. We propose μSwitch, an efficient kernel context isolation mechanism with memory protection that addresses these limitations. We use a protected structure, shared by the kernel and the user space, for context switching and propose implicit context switching to improve its performance by deferring the kernel resource switch to the next system call. We apply μSWITCH to isolate libraries in the Firefox web browser and an HTTP server, and reduce the overhead of isolation by 32.7% to 98.4% compared with other isolation techniques.more » « less
-
Finding the mode of a high dimensional probability distribution $$\mathcal{D}$$ is a fundamental algorithmic problem in statistics and data analysis. There has been particular interest in efficient methods for solving the problem when $$\mathcal{D}$$ is represented as a mixture model or kernel density estimate, although few algorithmic results with worst-case approximation and runtime guarantees are known. In this work, we significantly generalize a result of (LeeLiMusco:2021) on mode approximation for Gaussian mixture models. We develop randomized dimensionality reduction methods for mixtures involving a broader class of kernels, including the popular logistic, sigmoid, and generalized Gaussian kernels. As in Lee et al.’s work, our dimensionality reduction results yield quasi-polynomial algorithms for mode finding with multiplicative accuracy $$(1-\epsilon)$$ for any $$\epsilon > 0$$. Moreover, when combined with gradient descent, they yield efficient practical heuristics for the problem. In addition to our positive results, we prove a hardness result for box kernels, showing that there is no polynomial time algorithm for finding the mode of a kernel density estimate, unless $$\mathit{P} = \mathit{NP}$$. Obtaining similar hardness results for kernels used in practice (like Gaussian or logistic kernels) is an interesting future direction.more » « less
-
A recent technology breakthrough in spatial molecular profiling (SMP) has enabled the comprehensive molecular characterizations of single cells while preserving spatial information. It provides new opportunities to delineate how cells from different origins form tissues with distinctive structures and functions. One immediate question in SMP data analysis is to identify genes whose expressions exhibit spatially correlated patterns, called spatially variable (SV) genes. Most current methods to identify SV genes are built upon the geostatistical model with Gaussian process to capture the spatial patterns. However, the Gaussian process models rely on ad hoc kernels that could limit the models' ability to identify complex spatial patterns. In order to overcome this challenge and capture more types of spatial patterns, we introduce a Bayesian approach to identify SV genes via a modified Ising model. The key idea is to use the energy interaction parameter of the Ising model to characterize spatial expression patterns. We use auxiliary variable Markov chain Monte Carlo algorithms to sample from the posterior distribution with an intractable normalizing constant in the model. Simulation studies using both simulated and synthetic data showed that the energy‐based modeling approach led to higher accuracy in detecting SV genes than those kernel‐based methods. When applied to two real spatial transcriptomics (ST) datasets, the proposed method discovered novel spatial patterns that shed light on the biological mechanisms. In summary, the proposed method presents a new perspective for analyzing ST data.more » « less
-
Abstract Although prior studies have evaluated the role of sampling errors associated with local and regional methods to estimate peak flow quantiles, the investigation of epistemic errors is more difficult because the underlying properties of the random variable have been prescribed using ad‐hoc characterizations of the regional distributions of peak flows. This study addresses this challenge using representations of regional peak flow distributions derived from a combined framework of stochastic storm transposition, radar rainfall observations, and distributed hydrologic modeling. The authors evaluated four commonly used peak flow quantile estimation methods using synthetic peak flows at 5,000 sites in the Turkey River watershed in Iowa, USA. They first used at‐site flood frequency analysis using the Pearson Type III distribution with L‐moments. The authors then pooled regional information using (1) the index flood method, (2) the quantile regression technique, and (3) the parameter regression. This approach allowed quantification of error components stemming from epistemic assumptions, parameter estimation method, sample size, and, in the regional approaches, the number ofpooledsites. The results demonstrate that the inability to capture the spatial variability of the skewness of the peak flows dominates epistemic error for regional methods. We concluded that, in the study basin, this variability could be partially explained by river network structure and the predominant orientation of the watershed. The general approach used in this study is promising in that it brings new tools and sources of data to the study of the old hydrologic problem of flood frequency analysis.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
