arXiv subjects

Ling Zhan

Publications and source records attributed to Ling Zhan.

7 recordsLinked to original sources

arXiv2026

Position: AI for Science Should Treat Measurement-to-Dataset Pipelines as Inference Components

AI for Science (AI4Science) workflows often treat the released dataset as a fixed interface to the underlying system. However, in domains relying on \emph{indirect observation}, the learner observes a derivative representation produced by multi-stage measurement, reconstruction, and preprocessing pipelines. \textbf{We argue that these measurement-to-dataset pipelines are inference components: treating their outputs as ``given data'' freezes an observation model and obscures uncertainty over feasible pipeline choices.} We identify three failure modes arising from this ``frozen lens'': \textbf{(C1) hidden hypothesis space}, where the released dataset does not specify the pipeline configuration or its validity conditions; \textbf{(C2) uncertified transportability}, where a pipeline may be documented but its regime of validity is untested, so failures under distribution shift cannot be adjudicated; \textbf{(C3) ungoverned multiplicity}, where many defensible pipelines exist and dispersion is real but not propagated into uncertainty-aware evidence. We stress-test these claims with a large-scale neuroscience empirical audit, finding a survival rate of $\approx 0.0004\%$ under a cross-dataset stability criterion. We call on the AI4Science community to make pipelines \emph{computable} inference objects via domain-specific Computable Observation Frameworks. This shift enables quantifying pipeline adequacy and stability, converting implicit implementation choices into auditable, reproducible, and cumulative scientific evidence.

cs.LG↗

arXiv2026

Accelerating Benchmarking of Functional Connectivity Modeling via Structure-aware Core-set Selection

Benchmarking the hundreds of functional connectivity (FC) modeling methods on large-scale fMRI datasets is critical for reproducible neuroscience. However, the combinatorial explosion of model-data pairings makes exhaustive evaluation computationally prohibitive, preventing such assessments from becoming a routine pre-analysis step. To break this bottleneck, we reframe the challenge of FC benchmarking by selecting a small, representative core-set whose sole purpose is to preserve the relative performance ranking of FC operators. We formalize this as a ranking-preserving subset selection problem and propose Structure-aware Contrastive Learning for Core-set Selection (SCLCS), a self-supervised framework to select these core-sets. SCLCS first uses an adaptive Transformer to learn each sample's unique FC structure. It then introduces a novel Structural Perturbation Score (SPS) to quantify the stability of these learned structures during training, identifying samples that represent foundational connectivity archetypes. Finally, while SCLCS identifies stable samples via a top-k ranking, we further introduce a density-balanced sampling strategy as a necessary correction to promote diversity, ensuring the final core-set is both structurally robust and distributionally representative. On the large-scale REST-meta-MDD dataset, SCLCS preserves the ground-truth model ranking with just 10% of the data, outperforming state-of-the-art (SOTA) core-set selection methods by up to 23.2% in ranking consistency (nDCG@k). To our knowledge, this is the first work to formalize core-set selection for FC operator benchmarking, thereby making large-scale operators comparisons a feasible and integral part of computational neuroscience. Code is publicly available on https://github.com/lzhan94swu/SCLCS

cs.LG↗

arXiv2025

Beyond Pairwise Connections: Extracting High-Order Functional Brain Network Structures under Global Constraints

Functional brain network (FBN) modeling often relies on local pairwise interactions, whose limitation in capturing high-order dependencies is theoretically analyzed in this paper. Meanwhile, the computational burden and heuristic nature of current hypergraph modeling approaches hinder end-to-end learning of FBN structures directly from data distributions. To address this, we propose to extract high-order FBN structures under global constraints, and implement this as a Global Constraints oriented Multi-resolution (GCM) FBN structure learning framework. It incorporates 4 types of global constraint (signal synchronization, subject identity, expected edge numbers, and data labels) to enable learning FBN structures for 4 distinct levels (sample/subject/group/project) of modeling resolution. Experimental results demonstrate that GCM achieves up to a 30.6% improvement in relative accuracy and a 96.3% reduction in computational time across 5 datasets and 2 task settings, compared to 9 baselines and 10 state-of-the-art methods. Extensive experiments validate the contributions of individual components and highlight the interpretability of GCM. This work offers a novel perspective on FBN structure learning and provides a foundation for interdisciplinary applications in cognitive neuroscience. Code is publicly available on https://github.com/lzhan94swu/GCM.

cs.LG↗

arXiv2024

Quantifying sex differences in brain network topology by aggregating nodal centrality rankings

Although numerous studies report significant sex differences in functional connectivity, these differences do not sufficient to reveal specific functional disparities among brain regions or the topological differences in brain networks. Meanwhile, individual differences could potentially bias the understanding of these sex differences. To address these challenges, we propose a consensus rank-based method to quantify sex differences in four node centrality ranking within the functional brain network. This method aggregates individuals' nodal centrality rankings into a consensus or "average" ranking, minimizing the impact of outliers and enhancing the robustness of the findings. By analyzing resting-state functional MRI data from 1,948 healthy young adults (aged 18-27 years, 1,163 females), we find significant sex differences in the topology of functional brain network, primarily attributed to biological sex rather than individual differences. Particularly, sex accounts for approximately 10% of the differences in nodal centrality consensus rankings. Using a rank difference index (RDI), we identify eight critical brain regions with the greatest rank differences, including the insula, supramarginal gyrus, and dorsolateral superior frontal gyrus. females show higher rankings in regions with stronger intra-system connections, whereas males dominate in areas with stronger inter-system connections. Our findings enhance our understanding of sex-specific characteristics in functional brain networks. Moreover, our approach may offer novel insights into targeted population studies, including those involving healthy individuals and patients with brain injuries.

q-bio.QM↗

arXiv2024

Sex Differences in Hierarchical and Modular Organization of Functional Brain Networks: Insights from Hierarchical Entropy and Modularity Analysis

Existing studies have demonstrated significant sex differences in the neural mechanisms of daily life and neuropsychiatric disorders. The hierarchical organization of the functional brain network is a critical feature for assessing these neural mechanisms. But sex differences on the hierarchical organization is not fully investigated. Here, we explore whether hierarchical structure of brain network differs between females and males. At the group level, we measure the hierarchical entropy and the maximum modularity of each individual, and identify a significant negative correlation between the complexity of hierarchy and modularity in brain networks. On average, female brain networks have stronger connectivity within the module, whereas male brain networks demonstrate more complex hierarchy. At the consensus level, we use co-classification matrix to investigate the detailed differences in the hierarchical organization between sexes and observe that females and males exhibit different interaction patterns of brain regions in the dorsal attention network (DAN) and visual network (VIN). Our finding suggests that females and males employ different network topology to achieve brain functions. In addition, the negative correlation between hierarchy and modularity implies a need to balance the complexity in hierarchical organization of the brain network, which shed light on future studies of brain functions.

q-bio.NC↗

arXiv2023

Multi-feature concatenation and multi-classifier stacking: an interpretable and generalizable machine learning method for MDD discrimination with rsfMRI

Major depressive disorder is a serious and heterogeneous psychiatric disorder that needs accurate diagnosis. Resting-state functional MRI (rsfMRI), which captures multiple perspectives on brain structure, function, and connectivity, is increasingly applied in the diagnosis and pathological research of mental diseases. Different machine learning algorithms are then developed to exploit the rich information in rsfMRI and discriminate MDD patients from normal controls. Despite recent advances reported, the discrimination accuracy has room for further improvement. The generalizability and interpretability of the method are not sufficiently addressed either. Here, we propose a machine learning method (MFMC) for MDD discrimination by concatenating multiple features and stacking multiple classifiers. MFMC is tested on the REST-meta-MDD data set that contains 2428 subjects collected from 25 different sites. MFMC yields 96.9% MDD discrimination accuracy, demonstrating a significant improvement over existing methods. In addition, the generalizability of MFMC is validated by the good performance when the training and testing subjects are from independent sites. The use of XGBoost as the meta classifier allows us to probe the decision process of MFMC. We identify 13 feature values related to 9 brain regions including the posterior cingulate gyrus, superior frontal gyrus orbital part, and angular gyrus, which contribute most to the classification and also demonstrate significant differences at the group level. The use of these 13 feature values alone can reach 87% of MFMC's full performance when taking all feature values. These features may serve as clinically useful diagnostic and prognostic biomarkers for mental disorders in the future.

cs.LG↗

arXiv2022

CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling

Heterogeneous information network (HIN) embedding aims to find the representations of nodes that preserve the proximity between entities of different nature. A family of approaches that are wildly adopted applies random walk to generate a sequence of heterogeneous context, from which the embedding is learned. However, due to the multipartite graph structure of HIN, hub nodes tend to be over-represented in the sampled sequence, giving rise to imbalanced samples of the network. Here we propose a new embedding method CoarSAS2hvec. The self-avoid short sequence sampling with the HIN coarsening procedure (CoarSAS) is utilized to better collect the rich information in HIN. An optimized loss function is used to improve the performance of the HIN structure embedding. CoarSAS2hvec outperforms nine other methods in two different tasks on four real-world data sets. The ablation study confirms that the samples collected by CoarSAS contain richer information of the network compared with those by other methods, which is characterized by a higher information entropy. Hence, the traditional loss function applied to samples by CoarSAS can also yield improved results. Our work addresses a limitation of the random-walk-based HIN embedding that has not been emphasized before, which can shed light on a range of problems in HIN analyses.

cs.LG↗