... International Conference on Learning Representations最新文献_第4页

Probabilistic Geometric Principal Component Analysis with application to neural data. 概率几何主成分分析及其在神经数据中的应用。

... International Conference on Learning Representations Pub Date : 2025-04-01

Han-Lin Hsieh, Maryam M Shanechi

{"title":"Probabilistic Geometric Principal Component Analysis with application to neural data.","authors":"Han-Lin Hsieh, Maryam M Shanechi","doi":"","DOIUrl":"","url":null,"abstract":"Dimensionality reduction is critical across various domains of science including neuroscience. Probabilistic Principal Component Analysis (PPCA) is a prominent dimensionality reduction method that provides a probabilistic approach unlike the deterministic approach of PCA and serves as a connection between PCA and Factor Analysis (FA). Despite their power, PPCA and its extensions are mainly based on linear models and can only describe the data in a Euclidean coordinate system around the mean of data. However, in many neuroscience applications, data may be distributed around a nonlinear geometry (i.e., manifold) rather than lying in the Euclidean space around the mean. We develop Probabilistic Geometric Principal Component Analysis (PGPCA) for such datasets as a new dimensionality reduction algorithm that can explicitly incorporate knowledge about a given nonlinear manifold that is first fitted from these data. Further, we show how in addition to the Euclidean coordinate system, a geometric coordinate system can be derived for the manifold to capture the deviations of data from the manifold and noise. We also derive a data-driven EM algorithm for learning the PGPCA model parameters. As such, PGPCA generalizes PPCA to better describe data distributions by incorporating a nonlinear manifold geometry. In simulations and brain data analyses, we show that PGPCA can effectively model the data distribution around various given manifolds and outperforms PPCA for such data. Moreover, PGPCA provides the capability to test whether the new geometric coordinate system better describes the data than the Euclidean one. Finally, PGPCA can perform dimensionality reduction and learn the data distribution both around and on the manifold. These capabilities make PGPCA valuable for enhancing the efficacy of dimensionality reduction for analysis of high-dimensional data that exhibit noise and are distributed around a nonlinear manifold, especially for neural data.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"58365-58383"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13166129/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147937250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MediConfusion: Can you trust your AI radiologist? Probing the reliability of multimodal medical foundation models. MediConfusion：你能相信你的人工智能放射科医生吗？探讨多模态医学基础模型的可靠性。

... International Conference on Learning Representations Pub Date : 2025-04-01

Mohammad Shahab Sepehri, Zalan Fabian, Maryam Soltanolkotabi, Mahdi Soltanolkotabi

{"title":"MediConfusion: Can you trust your AI radiologist? Probing the reliability of multimodal medical foundation models.","authors":"Mohammad Shahab Sepehri, Zalan Fabian, Maryam Soltanolkotabi, Mahdi Soltanolkotabi","doi":"","DOIUrl":"","url":null,"abstract":"Multimodal Large Language Models (MLLMs) have tremendous potential to improve the accuracy, availability, and cost-effectiveness of healthcare by providing automated solutions or serving as aids to medical professionals. Despite promising first steps in developing medical MLLMs in the past few years, their capabilities and limitations are not well understood. Recently, many benchmark datasets have been proposed that test the general medical knowledge of such models across a variety of medical areas. However, the systematic failure modes and vulnerabilities of such models are severely underexplored with most medical benchmarks failing to expose the shortcomings of existing models in this safety-critical domain. In this paper, we introduce MediConfusion, a challenging medical Visual Question Answering (VQA) benchmark dataset, that probes the failure modes of medical MLLMs from a vision perspective. We reveal that state-of-the-art models are easily confused by image pairs that are otherwise visually dissimilar and clearly distinct for medical experts. Strikingly, all available models (open-source or proprietary) achieve performance below random guessing on MediConfusion, raising serious concerns about the reliability of existing medical MLLMs for healthcare deployment. We also extract common patterns of model failure that may help the design of a new generation of more trustworthy and reliable MLLMs in healthcare. The dataset and evaluation code are available at https://github.com/AIF4S/MediConfusion.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"41122-41146"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12915564/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146230540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Time-to-Event Pretraining for 3D Medical Imaging. 三维医学成像的事件时间预训练。

... International Conference on Learning Representations Pub Date : 2025-04-01

Zepeng Huo, Jason Alan Fries, Alejandro Lozano, Jeya Maria Jose Valanarasu, Ethan Steinberg, Louis Blankemeier, Akshay S Chaudhari, Curtis Langlotz, Nigam H Shah

{"title":"Time-to-Event Pretraining for 3D Medical Imaging.","authors":"Zepeng Huo, Jason Alan Fries, Alejandro Lozano, Jeya Maria Jose Valanarasu, Ethan Steinberg, Louis Blankemeier, Akshay S Chaudhari, Curtis Langlotz, Nigam H Shah","doi":"","DOIUrl":"","url":null,"abstract":"With the rise of medical foundation models and the growing availability of imaging data, scalable pretraining techniques offer a promising way to identify imaging biomarkers predictive of future disease risk. While current self-supervised methods for 3D medical imaging models capture local structural features like organ morphology, they fail to link pixel biomarkers with long-term health outcomes due to a missing context problem. Current approaches lack the temporal context necessary to identify biomarkers correlated with disease progression, as they rely on supervision derived only from images and concurrent text descriptions. To address this, we introduce time-to-event pretraining, a pretraining framework for 3D medical imaging models that leverages large-scale temporal supervision from paired, longitudinal electronic health records (EHRs). Using a dataset of 18,945 CT scans (4.2 million 2D images) and time-to-event distributions across thousands of EHR-derived tasks, our method improves outcome prediction, achieving an average AUROC increase of 23.7% and a 29.4% gain in Harrell's C-index across 8 benchmark tasks. Importantly, these gains are achieved without sacrificing diagnostic classification performance. This study lays the foundation for integrating longitudinal EHR and 3D imaging data to advance clinical risk prediction.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"100815-100851"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12950328/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147346152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hierarchical uncertainty estimation for learning-based registration in neuroimaging. 神经影像学中基于学习的配准的层次不确定性估计。

... International Conference on Learning Representations Pub Date : 2025-04-01

Xiaoling Hu, Karthik Gopinath, Peirong Liu, Malte Hoffmann, Koen Van Leemput, Oula Puonti, Juan Eugenio Iglesias

{"title":"Hierarchical uncertainty estimation for learning-based registration in neuroimaging.","authors":"Xiaoling Hu, Karthik Gopinath, Peirong Liu, Malte Hoffmann, Koen Van Leemput, Oula Puonti, Juan Eugenio Iglesias","doi":"","DOIUrl":"","url":null,"abstract":"Over recent years, deep learning based image registration has achieved impressive accuracy in many domains, including medical imaging and, specifically, human neuroimaging with magnetic resonance imaging (MRI). However, the uncertainty estimation associated with these methods has been largely limited to the application of generic techniques (e.g., Monte Carlo dropout) that do not exploit the peculiarities of the problem domain, particularly spatial modeling. Here, we propose a principled way to propagate uncertainties (epistemic or aleatoric) estimated at the level of spatial location by these methods, to the level of global transformation models, and further to downstream tasks. Specifically, we justify the choice of a Gaussian distribution for the local uncertainty modeling, and then propose a framework where uncertainties spread across hierarchical levels, depending on the choice of transformation model. Experiments on publicly available data sets show that Monte Carlo dropout correlates very poorly with the reference registration error, whereas our uncertainty estimates correlate much better. Crucially, the results also show that uncertainty-aware fitting of transformations improves the registration accuracy of brain MRI scans. Finally, we illustrate how sampling from the posterior distribution of the transformations can be used to propagate uncertainties to downstream neuroimaging tasks. Code is available at: https://github.com/HuXiaoling/Regre4Regis.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"91200-91216"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12943796/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147328781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Linear Spherical Sliced Optimal Transport: A Fast Metric for Comparing Spherical Data. 线性球面切片最优传输：球面数据比较的快速度量。

... International Conference on Learning Representations Pub Date : 2025-04-01

Xinran Liu, Yikun Bai, Rocío Díaz Martín, Kaiwen Shi, Ashkan Shahbazi, Bennett A Landman, Catie Chang, Soheil Kolouri

{"title":"Linear Spherical Sliced Optimal Transport: A Fast Metric for Comparing Spherical Data.","authors":"Xinran Liu, Yikun Bai, Rocío Díaz Martín, Kaiwen Shi, Ashkan Shahbazi, Bennett A Landman, Catie Chang, Soheil Kolouri","doi":"","DOIUrl":"","url":null,"abstract":"Efficient comparison of spherical probability distributions becomes important in fields such as computer vision, geosciences, and medicine. Sliced optimal transport distances, such as spherical and stereographic spherical sliced Wasserstein distances, have recently been developed to address this need. These methods reduce the computational burden of optimal transport by slicing hyperspheres into one-dimensional projections, i.e., lines or circles. Concurrently, linear optimal transport has been proposed to embed distributions into <math> <mrow><msup><mi>L</mi> <mn>2</mn></msup> </mrow> </math> spaces, where the <math> <mrow><msup><mi>L</mi> <mn>2</mn></msup> </mrow> </math> distance approximates the optimal transport distance, thereby simplifying comparisons across multiple distributions. In this work, we introduce the Linear Spherical Sliced Optimal Transport (LSSOT) framework, which utilizes slicing to embed spherical distributions into <math> <mrow><msup><mi>L</mi> <mn>2</mn></msup> </mrow> </math> spaces while preserving their intrinsic geometry, offering a computationally efficient metric for spherical probability measures. We establish the metricity of LSSOT and demonstrate its superior computational efficiency in applications such as cortical surface registration, 3D point cloud interpolation via gradient flow, and shape embedding. Our results demonstrate the significant computational benefits and high accuracy of LSSOT in these applications.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"35195-35237"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13051564/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147635663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

GPS: A Probabilistic Distributional Similarity with Gumbel Priors for Set-to-Set Matching. GPS：具有Gumbel先验的集对集匹配的概率分布相似度。

... International Conference on Learning Representations Pub Date : 2025-04-01

Ziming Zhang, Fangzhou Lin, Haotian Liu, Jose Morales, Haichong Zhang, Kazunori Yamada, Vijaya B Kolachalama, Venkatesh Saligrama

{"title":"GPS: A Probabilistic Distributional Similarity with Gumbel Priors for Set-to-Set Matching.","authors":"Ziming Zhang, Fangzhou Lin, Haotian Liu, Jose Morales, Haichong Zhang, Kazunori Yamada, Vijaya B Kolachalama, Venkatesh Saligrama","doi":"","DOIUrl":"","url":null,"abstract":"Set-to-set matching aims to identify correspondences between two sets of unordered items by minimizing a distance metric or maximizing a similarity measure. Traditional metrics, such as Chamfer Distance (CD) and Earth Mover's Distance (EMD), are widely used for this purpose but often suffer from limitations like suboptimal performance in terms of accuracy and robustness, or high computational costs - or both. In this paper, we propose an effective set-to-set matching similarity measure, GPS, based on Gumbel prior distributions. These distributions are typically used to model the extrema of samples drawn from various distributions. Our approach is motivated by the observation that the distributions of minimum distances from CD, as encountered in real-world applications such as point cloud completion, can be accurately modeled using Gumbel distributions. We validate our method on tasks like few-shot image classification and 3D point cloud completion, demonstrating significant improvements over state-of-the-art loss functions across several benchmark datasets. Our demo code is publicly available at https://github.com/Zhang-VISLab/ICLR2025-GPS.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"84407-84421"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13135232/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147826387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

BRAID: Input-driven nonlinear dynamical modeling of neural-behavioral data. 神经行为数据的输入驱动非线性动态建模。

... International Conference on Learning Representations Pub Date : 2025-04-01

Parsa Vahidi, Omid G Sani, Maryam M Shanechi

{"title":"BRAID: Input-driven nonlinear dynamical modeling of neural-behavioral data.","authors":"Parsa Vahidi, Omid G Sani, Maryam M Shanechi","doi":"","DOIUrl":"","url":null,"abstract":"Neural populations exhibit complex recurrent structures that drive behavior, while continuously receiving and integrating external inputs from sensory stimuli, upstream regions, and neurostimulation. However, neural populations are often modeled as autonomous dynamical systems, with little consideration given to the influence of external inputs that shape the population activity and behavioral outcomes. Here, we introduce BRAID, a deep learning framework that models nonlinear neural dynamics underlying behavior while explicitly incorporating any measured external inputs. Our method disentangles intrinsic recurrent neural population dynamics from the effects of inputs by including a forecasting objective within input-driven recurrent neural networks. BRAID further prioritizes the learning of intrinsic dynamics that are related to a behavior of interest by using a multi-stage optimization scheme. We validate BRAID with nonlinear simulations, showing that it can accurately learn the intrinsic dynamics shared between neural and behavioral modalities. We then apply BRAID to motor cortical activity recorded during a motor task and demonstrate that our method more accurately fits the neural-behavioral data by incorporating measured sensory stimuli into the model and improves the forecasting of neural-behavioral data compared with various baseline methods, whether input-driven or not.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"6610-6639"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13166130/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147937222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Continual Slow-and-Fast Adaptation of Latent Neural Dynamics (CoSFan): Meta-Learning What-How & When to Adapt. 潜在神经动力学（CoSFan）的持续慢速和快速适应：元学习什么-如何适应和何时适应。

... International Conference on Learning Representations Pub Date : 2025-04-01

Ryan Missel, Linwei Wang

{"title":"Continual Slow-and-Fast Adaptation of Latent Neural Dynamics (CoSFan): Meta-Learning What-How & When to Adapt.","authors":"Ryan Missel, Linwei Wang","doi":"","DOIUrl":"","url":null,"abstract":"An increasing interest in learning to forecast for time-series of high-dimensional observations is the ability to adapt to systems with diverse underlying dynamics. Access to observations that define a stationary distribution of these systems is often unattainable, as the underlying dynamics may change over time. Naively training or retraining models at each shift may lead to catastrophic forgetting about previously-seen systems. We present a new continual meta-learning (CML) framework to realize continual slow-and fast adaptation of latent dynamics (CoSFan). We leverage a feed-forward meta-model to infer what the current system is and how to adapt a latent dynamics function to it, enabling fast adaptation to specific dynamics. We then develop novel strategies to automatically detect when a shift of data distribution occurs, with which to identify its underlying dynamics and its relation with previously-seen dynamics. In combination with fixed-memory experience replay mechanisms, this enables continual slow update of the what-how meta-model. Empirical studies demonstrated that both the meta- and continual-learning component was critical for learning to forecast across non-stationary distributions of diverse dynamics systems, and the feed-forward meta-model combined with task-aware/-relational continual learning strategies significantly outperformed existing CML alternatives.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"29253-29286"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13284768/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148311301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Global Identifiability of Overcomplete Dictionary Learning via L1 and Volume Minimization. 基于L1和体积最小化的过完全字典学习的全局可识别性。

... International Conference on Learning Representations Pub Date : 2025-01-01

Yuchen Sun, Kejun Huang

{"title":"Global Identifiability of Overcomplete Dictionary Learning via L1 and Volume Minimization.","authors":"Yuchen Sun, Kejun Huang","doi":"","DOIUrl":"","url":null,"abstract":"We propose a novel formulation for dictionary learning with an overcomplete dictionary, i.e., when the number of atoms is larger than the dimension of the dictionary. The proposed formulation consists of a weighted sum of <math> <msub><mrow><mi>ℓ</mi></mrow> <mrow><mn>1</mn></mrow> </msub> </math> norms of the rows of the sparse coefficient matrix plus the log of the matrix volume of the dictionary matrix. The main contribution of this work is to show that this novel formulation guarantees global identifiability of the overcomplete dictionary, under a mild condition that the sparse coefficient matrix satisfies a strong scattering condition in the hypercube. Furthermore, if every column of the coefficient matrix is sparse and the dictionary guarantees <math> <msub><mrow><mi>ℓ</mi></mrow> <mrow><mn>1</mn></mrow> </msub> </math> recovery, then the coefficient matrix is identifiable as well. This is a major breakthrough for not only dictionary learning but also general matrix factorization models as identifiability is guaranteed even when the latent dimension is higher than the ambient dimension. We also provide a probabilistic analysis and show that if the sparse coefficient matrix is generated from the widely adopted sparse-Gaussian model, then the <math><mi>m</mi> <mo>×</mo> <mi>k</mi></math> overcomplete dictionary is globally identifiable if the sample size is bigger than a constant times <math> <mfenced> <mrow> <msup><mrow><mi>k</mi></mrow> <mrow><mn>2</mn></mrow> </msup> <mo>/</mo> <mi>m</mi></mrow> </mfenced> <mtext>log</mtext> <mfenced> <mrow> <msup><mrow><mi>k</mi></mrow> <mrow><mn>2</mn></mrow> </msup> <mo>/</mo> <mi>m</mi></mrow> </mfenced> </math> with overwhelming probability. Finally, we propose an algorithm based on alternating minimization to solve the new proposed formulation.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"53462-53479"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13636083/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149123928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving Graph Neural Networks by Learning Continuous Edge Directions. 通过学习连续边缘方向改进图神经网络。

... International Conference on Learning Representations Pub Date : 2025-01-01

Seong Ho Pahng, Sahand Hormoz

{"title":"Improving Graph Neural Networks by Learning Continuous Edge Directions.","authors":"Seong Ho Pahng, Sahand Hormoz","doi":"","DOIUrl":"","url":null,"abstract":"Graph Neural Networks (GNNs) traditionally employ a message-passing mechanism that resembles diffusion over undirected graphs, which often leads to homogenization of node features and reduced discriminative power in tasks such as node classification. Our key insight for addressing this limitation is to assign fuzzy edge directions-that can vary continuously from node <math><mi>i</mi></math> pointing to node <math><mi>j</mi></math> to vice versa-to the edges of a graph so that features can preferentially flow in one direction between nodes to enable long-range information transmission across the graph. We also introduce a novel complex-valued Laplacian for directed graphs with fuzzy edges where the real and imaginary parts represent information flow in opposite directions. Using this Laplacian, we propose a general framework, called Continuous Edge Direction (CoED) GNN, for learning on graphs with fuzzy edges and prove its expressivity limits using a generalization of the Weisfeiler-Leman (WL) graph isomorphism test for directed graphs with fuzzy edges. Our architecture aggregates neighbor features scaled by the learned edge directions and processes the aggregated messages from in-neighbors and out-neighbors separately alongside the self-features of the nodes. Since continuous edge directions are differentiable, they can be learned jointly with the GNN weights via gradient-based optimization. CoED GNN is particularly well-suited for graph ensemble data where the graph structure remains fixed but multiple realizations of node features are available, such as in gene regulatory networks, web connectivity graphs, and power grids. We demonstrate through extensive experiments on both synthetic and real graph ensemble datasets that learning continuous edge directions significantly improves performance both for undirected and directed graphs compared with existing methods. Our code is available on GitHub.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2025 ","pages":"46849-46884"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13592212/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149019570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0