arXiv subjects

Yuefan Deng

Publications and source records attributed to Yuefan Deng.

18 recordsLinked to original sources

arXiv2026

HP-JEPA: Hierarchical Partitioning for Multi-Resolution Graph Joint-Embedding Predictive Learning

Graph self-supervised learning aims to learn transferable representations from large-scale unlabeled graph data. Joint-embedding predictive architectures (JEPAs) avoid explicit negative-pair construction and raw-input reconstruction by predicting masked targets directly in latent space. However, existing graph JEPAs typically rely on a single predefined graph partition, biasing the learned representations toward one structural granularity and limiting their ability to capture complementary patterns at different graph scales. To address this limitation, we propose HP-JEPA, a hierarchical partitioning framework for multi-resolution graph joint-embedding prediction. HP-JEPA organizes each graph into an ordered bank of coarse-to-fine partition resolutions and performs context-target latent prediction separately at each resolution using an online encoder, an exponential-moving-average target encoder, and a latent predictor. The resulting resolution-specific graph representations are subsequently integrated through concatenation or task-specific resolution weighting, allowing downstream models to combine complementary local, regional, and global structural information. Experiments on seven graph classification benchmarks and one graph regression benchmark show that HP-JEPA outperforms the fixed-resolution Graph-JEPA baseline on 6 of 8 tasks, improving upon Graph-JEPA on most evaluated benchmarks. Size-stratified analyses further show that HP-JEPA achieves higher accuracy than Graph-JEPA in most evaluated graph-size quartiles on three representative datasets. These results highlight the effectiveness of hierarchical multi-resolution partitioning for transferable graph representation learning.

cs.LG↗

arXiv2026

A New Broadcast Model for Several Network Topologies

We introduce Broadcast by Balanced Saturation (BBS), a general class of tree-based pipelined broadcast algorithms that optimizes communication efficiency across diverse network topologies, with a particular emphasis on large message sizes. By addressing spanning tree construction and communication task scheduling, two fundamental theoretical challenges in broadcasting, BBS offers a unified and flexible framework that operates effectively under varied network constraints. The algorithm maximizes aggregated throughput while simultaneously addressing topology constraints, synchronization overhead, bandwidth limitations and contention. Using SimGrid under standard assumptions, including full-duplex and one-port communication, various algorithms were evaluated on Mesh, Butterfly, Dragonfly, and Fat-Tree topologies. Results demonstrate that BBS consistently outperforms both general-purpose and topology-aware broadcast algorithms across a wide range of topologies and message sizes, establishing it as a robust and high-performance solution for large-scale systems.

cs.NI↗

arXiv2026

Towards Principled Test-Time Adaptation for Time Series Forecasting

Test-time adaptation (TTA) has recently emerged as a promising approach for improving time series forecasting (TSF) under distribution shift. Existing TSF-TTA methods differ in how they utilize revealed targets, yet the resulting adaptation protocols remain heterogeneous and lack a clearly unified formulation. To address this issue, we revisit TSF-TTA from the perspective of protocol cleanliness and propose an adaptation protocol based solely on matured ground truth, yielding a more principled setting for adaptation. Under this protocol, we further diagnose existing adapters in the frequency domain and find that their prediction corrections often exhibit limited and weakly structured spectral modifications. Motivated by this diagnosis, we propose Frequency-Aware Calibration (FAC), a lightweight calibration method that directly parameterizes prediction corrections in the frequency domain. Across diverse datasets, forecasting horizons, and source forecasters, FAC achieves competitive and consistent performance while requiring substantially fewer trainable parameters than the compared TSF-TTA adapters.

cs.LG↗

arXiv2025

Boundary-Informed Method of Lines for Physics Informed Neural Networks

We propose a hybrid solver that fuses the dimensionality-reduction strengths of the Method of Lines (MOL) with the flexibility of Physics-Informed Neural Networks (PINNs). Instead of approximating spatial derivatives with fixed finite-difference stencils - whose truncation errors force extremely fine meshes - our method trains a neural network to represent the initial spatial profile and then employs automatic differentiation to obtain spectrally accurate gradients at arbitrary nodes. These high-fidelity derivatives define the right-hand side of the MOL-generated ordinary-differential system, and time integration is replaced with a secondary temporal PINN while spatial accuracy is retained without mesh refinement. The resulting "boundary-informed MOL-PINN" matches or surpasses conventional MOL in accuracy using an order of magnitude fewer collocation points, thereby shrinking memory footprints, lessening dependence on large data sets, and increasing complexity robustness. Because it relies only on automatic differentiation and standard optimizers, the framework extends naturally to linear and nonlinear PDEs in any spatial dimension.

physics.comp-ph↗

arXiv2025

Kolmogorov-Arnold Representation for Symplectic Learning: Advancing Hamiltonian Neural Networks

We propose a Kolmogorov-Arnold Representation-based Hamiltonian Neural Network (KAR-HNN) that replaces the Multilayer Perceptrons (MLPs) with univariate transformations. While Hamiltonian Neural Networks (HNNs) ensure energy conservation by learning Hamiltonian functions directly from data, existing implementations, often relying on MLPs, cause hypersensitivity to the hyperparameters while exploring complex energy landscapes. Our approach exploits the localized function approximations to better capture high-frequency and multi-scale dynamics, reducing energy drift and improving long-term predictive stability. The networks preserve the symplectic form of Hamiltonian systems, and thus maintain interpretability and physical consistency. After assessing KAR-HNN on four benchmark problems including spring-mass, simple pendulum, two- and three-body problem, we foresee its effectiveness for accurate and stable modeling of realistic physical processes often at high dimensions and with few known parameters.

cs.LG↗

arXiv2025

An Iterative Framework for Generative Backmapping of Coarse Grained Proteins

The techniques of data-driven backmapping from coarse-grained (CG) to fine-grained (FG) representation often struggle with accuracy, unstable training, and physical realism, especially when applied to complex systems such as proteins. In this work, we introduce a novel iterative framework by using conditional Variational Autoencoders and graph-based neural networks, specifically designed to tackle the challenges associated with such large-scale biomolecules. Our method enables stepwise refinement from CG beads to full atomistic details. We outline the theory of iterative generative backmapping and demonstrate via numerical experiments the advantages of multistep schemes by applying them to proteins of vastly different structures with very coarse representations. This multistep approach not only improves the accuracy of reconstructions but also makes the training process more computationally efficient for proteins with ultra-CG representations.

cs.LG↗

arXiv2025

Velocity-Inferred Hamiltonian Neural Networks: Learning Energy-Conserving Dynamics from Position-Only Data

Data-driven modeling of physical systems often relies on learning both positions and momenta to accurately capture Hamiltonian dynamics. However, in many practical scenarios, only position measurements are readily available. In this work, we introduce a method to train a standard Hamiltonian Neural Network (HNN) using only position data, enabled by a theoretical result that permits transforming the Hamiltonian $H(q,p)$ into a form $H(q, v)$. Under certain assumptions, namely, an invertible relationship between momentum and velocity, we formally prove the validity of this substitution and demonstrate how it allows us to infer momentum from position alone. We apply our approach to canonical examples including the spring-mass system, pendulum, two-body, and three-body problems. Our results show that using only position data is sufficient for stable and energy-consistent long-term predictions, suggesting a promising pathway for data-driven discovery of Hamiltonian systems when momentum measurements are unavailable.

physics.comp-ph↗

arXiv2025

The Impact of Move Schemes on Simulated Annealing Performance

Designing an effective move-generation function for Simulated Annealing (SA) in complex models remains a significant challenge. In this work, we present a combination of theoretical analysis and numerical experiments to examine the impact of various move-generation parameters -- such as how many particles are moved and by what distance at each iteration -- under different temperature schedules and system sizes. Our numerical studies, carried out on both the Lennard-Jones problem and an additional benchmark, reveal that moving exactly one randomly chosen particle per iteration offers the most efficient performance. We analyze acceptance rates, exploration properties, and convergence behavior, providing evidence that partial-coordinate updates can outperform full-coordinate moves in certain high-dimensional settings. These findings offer practical guidelines for optimizing SA methods in a broad range of complex optimization tasks.

math.OC↗

arXiv2023

Exploring Robust Features for Improving Adversarial Robustness

While deep neural networks (DNNs) have revolutionized many fields, their fragility to carefully designed adversarial attacks impedes the usage of DNNs in safety-critical applications. In this paper, we strive to explore the robust features which are not affected by the adversarial perturbations, i.e., invariant to the clean image and its adversarial examples, to improve the model's adversarial robustness. Specifically, we propose a feature disentanglement model to segregate the robust features from non-robust features and domain specific features. The extensive experiments on four widely used datasets with different attacks demonstrate that robust features obtained from our model improve the model's adversarial robustness compared to the state-of-the-art approaches. Moreover, the trained domain discriminator is able to identify the domain specific features from the clean images and adversarial examples almost perfectly. This enables adversarial example detection without incurring additional computational costs. With that, we can also specify different classifiers for clean images and adversarial examples, thereby avoiding any drop in clean image accuracy.

cs.CV↗

arXiv2022

AI-aided multiscale modeling of physiologically-significant blood clots

We have developed an AI-aided multiple time stepping (AI-MTS) algorithm and multiscale modeling framework (AI-MSM) and implemented them on the Summit-like supercomputer, AIMOS. AI-MSM is the first of its kind to integrate multi-physics, including intra-platelet, inter-platelet, and fluid-platelet interactions, into one system. It has simulated a record-setting multiscale blood clotting model of 102 million particles, of which 70 flowing and 180 aggregating platelets, under dissipative particle dynamics to coarse-grained molecular dynamics. By adaptively adjusting timestep sizes to match the characteristic time scales of the underlying dynamics, AI-MTS optimally balances speeds and accuracies of the simulations.

physics.med-ph↗

arXiv2022

Optimal circulant graphs as low-latency network topologies

Communication latency has become one of the determining factors for the performance of parallel clusters. To design low-latency network topologies for high-performance computing clusters, we optimize the diameters, mean path lengths, and bisection widths of circulant topologies. We obtain a series of optimal circulant topologies of size $2^5$ through $2^{10}$ and compare them with torus and hypercube of the same sizes and degrees. We further benchmark on a broad variety of applications including effective bandwidth, FFTE, Graph 500 and NAS parallel benchmarks to compare the optimal circulant topologies and Cartesian products of optimal circulant topologies and fully connected topologies with corresponding torus and hypercube. Simulation results demonstrate superior potentials of the optimal circulant topologies for communication-intensive applications. We also find the strengths of the Cartesian products in exploiting global communication with data traffic patterns of specific applications or internal algorithms.

cs.NI↗

arXiv2021

AGKD-BML: Defense Against Adversarial Attack by Attention Guided Knowledge Distillation and Bi-directional Metric Learning

While deep neural networks have shown impressive performance in many tasks, they are fragile to carefully designed adversarial attacks. We propose a novel adversarial training-based model by Attention Guided Knowledge Distillation and Bi-directional Metric Learning (AGKD-BML). The attention knowledge is obtained from a weight-fixed model trained on a clean dataset, referred to as a teacher model, and transferred to a model that is under training on adversarial examples (AEs), referred to as a student model. In this way, the student model is able to focus on the correct region, as well as correcting the intermediate features corrupted by AEs to eventually improve the model accuracy. Moreover, to efficiently regularize the representation in feature space, we propose a bidirectional metric learning. Specifically, given a clean image, it is first attacked to its most confusing class to get the forward AE. A clean image in the most confusing class is then randomly picked and attacked back to the original class to get the backward AE. A triplet loss is then used to shorten the representation distance between original image and its AE, while enlarge that between the forward and backward AEs. We conduct extensive adversarial robustness experiments on two widely used datasets with different attacks. Our proposed AGKD-BML model consistently outperforms the state-of-the-art approaches. The code of AGKD-BML will be available at: https://github.com/hongw579/AGKD-BML.

cs.CV↗

arXiv2019

Load-balanced Routing for Nested Interconnection Networks

We introduced the load-balanced routing algorithms, for interconnection networks resulting from nesting, by considering the pressure of the data forwarding in each node. Benchmarks on a small cluster with various network topologies, and simulations for several larger clusters whose prototypes are too costly to construct, demonstrated substantial gains of communication performance with our routing on these networks over other mainstream routing algorithms.

cs.NI↗

arXiv2019

A Structured Table of Graphs with Symmetries and Other Special Properties

We organize a table of regular graphs with minimal diameters and minimal mean path lengths, large bisection widths and high degrees of symmetries, obtained by enumerations on supercomputers. These optimal graphs, many of which are newly discovered, may find wide applications, for example, in design of network topologies.

cs.DM↗

arXiv2019

A new record of graph enumeration enabled by parallel processing

Using three supercomputers, we broke a record set in 2011, in the enumeration of non-isomorphic regular graphs by expanding the sequence of A006820 in Online Encyclopedia of Integer Sequences (OEIS), to achieve the number for 4-regular graphs of order 23 as 429,668,180,677,439, while discovering serval optimal regular graphs with minimum average shortest path lengths (ASPL) that can be used as interconnection networks for parallel computers. The number of 4-regular graphs and the optimal graphs, extremely time-consuming to calculate, result from a method we adapt from GENREG, a classical regular graph generator, to fit for supercomputers' strengths of using thousands of processor cores.

cs.DM↗

arXiv2019

Optimal Low-Latency Network Topologies for Cluster Performance Enhancement

We propose that clusters interconnected with network topologies having minimal mean path length will increase their overall performance for a variety of applications. We approach our heuristic by constructing clusters of up to 36 nodes having Dragonfly, torus, ring, Chvatal, Wagner, Bidiakis and several other topologies with minimal mean path lengths and by simulating the performance of 256-node clusters with the same network topologies. The optimal (or sub-optimal) low-latency network topologies are found by minimizing the mean path length of regular graphs. The selected topologies are benchmarked using ping-pong messaging, the MPI collective communications, and the standard parallel applications including effective bandwidth, FFTE, Graph 500 and NAS parallel benchmarks. We established strong correlations between the clusters' performances and the network topologies, especially the mean path lengths, for a wide range of applications. In communication-intensive benchmarks, clusters with optimal network topologies out-perform those with mainstream topologies by several folds. It is striking that a mere adjustment of the network topology suffices to reclaim performance from the same computing hardware.

cs.NI↗

arXiv2017

Symmetry-guided design of topologies for supercomputer networks

A family of graphs optimized as the topologies for supercomputer interconnection networks is proposed. The special needs of such network topologies, minimal diameter and mean path length, are met by special constructions of the weight vectors in a representation of the symplectic algebra. Such theoretical design of topologies can conveniently reconstruct the mesh and hypercubic graphs, widely used as today's network topologies. Our symplectic algebraic approach helps generate many classes of graphs suitable for network topologies.

cs.DM↗

arXiv1998

Clustering Kinetics of Granular Media in Three Dimensions

Three-dimensional molecular dynamics simulations of dissipative particles (~ 10^6) are carried out for studying the clustering kinetics of granular media during cooling. The inter-connected high particle density regions are identified, showing tube-like structures. The energy decay rates as functions of the particle density and the restitution coefficient are obtained. It is found that the probability density function of the particle density obeys an exponential distribution at late stages. Both the fluctuation of density and the mean cluster size of the particle density have power law relations against time during the inelastic coalescing process.

cond-mat.mtrl-sci↗