arXiv subjects

Han Fan

Publications and source records attributed to Han Fan.

3 recordsLinked to original sources

arXiv2026

BrainTAP: Brain Disorder Prediction with Adaptive Distill and Selective Prior Integration

Predicting clinical outcomes from brain networks in large-scale neuroimaging cohorts such as the Adolescent Brain Cognitive Development (ABCD) study requires effectively integrating functional connectivity (FC) and structural connectivity (SC) while incorporating expert neurobiological knowledge. However, existing multimodal fusion approaches are shallow or over-homogenize the inherently heterogeneous characteristics of FC and SC, while expert-defined anatomical priors are underutilized with static integration. To address these limitations, we propose Brain Transformer with Adaptive Mutual-Distill and Selective Prior Fusion (BrainTAP). We introduce Adaptive Mutual Distill (AMD), which enables layer-wise information exchange between modalities through learnable distill-intact ratios, preserving modality-specific signals while capturing cross-modal synergies. We further develop Selective Prior Fusion (SPF), which integrates expert-defined anatomical priors in an adaptive way. Evaluated on the ABCD dataset for predicting attention-related disorders, BrainTAP achieves superior performance over state-of-the-art baselines, demonstrating its effectiveness for brain disorder prediction.

cs.CE↗

arXiv2024

LaCE-LHMP: Airflow Modelling-Inspired Long-Term Human Motion Prediction By Enhancing Laminar Characteristics in Human Flow

Long-term human motion prediction (LHMP) is essential for safely operating autonomous robots and vehicles in populated environments. It is fundamental for various applications, including motion planning, tracking, human-robot interaction and safety monitoring. However, accurate prediction of human trajectories is challenging due to complex factors, including, for example, social norms and environmental conditions. The influence of such factors can be captured through Maps of Dynamics (MoDs), which encode spatial motion patterns learned from (possibly scattered and partial) past observations of motion in the environment and which can be used for data-efficient, interpretable motion prediction (MoD-LHMP). To address the limitations of prior work, especially regarding accuracy and sensitivity to anomalies in long-term prediction, we propose the Laminar Component Enhanced LHMP approach (LaCE-LHMP). Our approach is inspired by data-driven airflow modelling, which estimates laminar and turbulent flow components and uses predominantly the laminar components to make flow predictions. Based on the hypothesis that human trajectory patterns also manifest laminar flow (that represents predictable motion) and turbulent flow components (that reflect more unpredictable and arbitrary motion), LaCE-LHMP extracts the laminar patterns in human dynamics and uses them for human motion prediction. We demonstrate the superior prediction performance of LaCE-LHMP through benchmark comparisons with state-of-the-art LHMP methods, offering an unconventional perspective and a more intuitive understanding of human movement patterns.

cs.RO↗

arXiv2014

How network temporal dynamics shape a mutualistic system with invasive species?

Ecological networks allow us to study the structure and function of ecosystems and gain insights on species resilience/stability. The study of this ecological networks is usually a snapshop focused in a limited specific range of space and time, prevent us to perceive the real dynamics of ecological processes. By definition, an alien species has some ecological strategies and traits that permit it to compete better than the native species (e.g. absence of predators, different bloom period, high grow rate, etc.). Plant-pollinator networks provide valuable services to whole ecosystems and the introduction of an alien species may have different effects on the native network (competitive facilitation, native species extinction, etc.). While scientists acknowledge the significance of network connectivity in driving ecosystem services, the inclusion of temporary networks in ecological models is still in its infancy. We propose to use existing data on seasonality to develop a simulation platform that show inference between temporality of networks and invasions traits. Our focus is only to pick up some simple model to show, that theoretically temporal aspect play a role (different extinction patterns) to encourage ecologist to get involved in temporal networks. Moreover, the derived simulations could be further extended and adjust to other ecological questions.

q-bio.PE↗