论文检索

Prabath Hetti Mudiyanselage, Ruwan Tennakoon, John Thangarajah, Robert Ware, Jason Callahan

Positron Emission Tomography (PET) is a powerful imaging modality for assessing biochemical processes within the body. However, accurate image reconstruction is challenged by photon attenuation, particularly in dense structures such as bones, leading to quantification errors and reduced diagnostic confidence. Computed Tomography (CT) based attenuation correction is the standard approach but introduces additional radiation exposure, longer imaging times, and patient inconvenience, as well as potential registration errors, motion artifacts, and energy scaling inaccuracies. In this study, we propose a 3D U-Net based deep learning framework that directly estimates attenuation information from PET sinograms, eliminating the need for additional imaging modalities. Our approach integrates PET physics and employs custom skip connections to enhance cross-domain learning. We evaluate our model on a simulated brain dataset derived from real patient templates, achieving a Dice coefficient of 0.650 and an accuracy of 0.486 for bone structures. The clinical applicability of our method is further assessed by reconstructing PET images with the generated attenuation maps, yielding an MSE of 0.007 and an SSIM of 0.956, demonstrating strong structural consistency with CT-based attenuation correction. These results highlight the feasibility of performing PET image attenuation correction using PET sinograms alone, offering a promising alternative that reduces imaging time, radiation exposure, and patient burden while enabling faster and more efficient PET reconstruction.

Bingke Zhu, Xiaoxiao Wang, Minghui Jia, Yihan Tao, Xiao Kong, Ali Luo, Yingying Chen, Ming Tang, Jinqiao Wang

Stellar flare events are critical observational samples for astronomical research; however, recorded flare events remain limited. Stellar flare forecasting can provide additional flare event samples to support research efforts. Despite this potential, no specialized models for stellar flare forecasting have been proposed to date. In this paper, we present extensive experimental evidence demonstrating that both stellar physical properties and historical flare records are valuable inputs for flare forecasting tasks. We then introduce FLARE (Forecasting Light-curve-based Astronomical Records via features Ensemble), the first-of-its-kind large model specifically designed for stellar flare forecasting. FLARE integrates stellar physical properties and historical flare records through a novel Soft Prompt Module and Residual Record Fusion Module. Experiments on the Kepler light curve dataset demonstrate that FLARE achieves superior performance compared to other methods across all evaluation metrics. Finally, we validate the forecast capability of our model through a comprehensive case study.

Jiaping Xiao, Rangya Zhang, Yuhang Zhang, Lu Bai, Qianlei Jia, Mir Feroskhan

Space situational awareness (SSA) plays an imperative role in maintaining safe space operations, especially given the increasingly congested space traffic around the Earth. Space-based SSA offers a flexible and lightweight solution compared to traditional ground-based SSA. With advanced machine learning approaches, space-based SSA can extract features from high-resolution images in space to detect and track resident space objects (RSOs). However, existing spacecraft image datasets, such as SPARK, fall short of providing realistic camera observations, rendering the derived algorithms unsuitable for real SSA systems. In this work, we introduce SpaceDet, a large-scale realistic space-based image dataset for SSA. We consider accurate space orbit dynamics and a physical camera model with various noise distributions, generating images at the photon level. To extend the available observation window, four overlapping cameras are simulated with a fixed rotation angle. SpaceDet includes images of RSOs observed from 19 km to 63,000 km, captured by a tracker operating in LEO, MEO, and GEO orbits over a period of 5,000 seconds. Each image has a resolution of 4418 x 4418 pixels, providing detailed features for developing advanced SSA approaches. We split the dataset into three subsets: SpaceDet-100, SpaceDet-5000, and SpaceDet-full, catering to various image processing applications. The SpaceDet-full corpus includes a comprehensive dataloader with 781.5 GB of images and 25.9 MB of ground truth labels. Furthermore, we adapted detection and tracking algorithms on the collected dataset using a specified splitting method to accelerate the training process. The trained model can detect RSOs from real-world space observations with zero-shot capability.

Satyavrat Wagle, Akshay Malhotra, Shahab Hamidi-Rad, Aditya Sant, David J. Love, Christopher G. Brinton

In recent years, machine learning (ML) methods have become increasingly popular in wireless communication systems for several applications. A critical bottleneck for designing ML systems for wireless communications is the availability of realistic wireless channel datasets, which are extremely resource-intensive to produce. To this end, the generation of realistic wireless channels plays a key role in the subsequent design of effective ML algorithms for wireless communication systems. Generative models have been proposed to synthesize channel matrices, but outputs produced by such methods may not correspond to geometrically viable channels and do not provide any insight into the scenario being generated. In this work, we aim to address both these issues by integrating established parametric, physics-based geometric channel (PPGC) modeling frameworks with generative methods to produce realistic channel matrices with interpretable representations in the parameter domain. We show that the generative model converges to prohibitively suboptimal stationary points when learning the underlying prior directly over the parameters due to the non-convex PPGC model. To address this limitation, we propose a linearized reformulation of the problem to ensure smooth gradient flow during generative model training, while also providing insights into the underlying physical environment. We evaluate our model against prior baselines by comparing the generated, scenario-specific samples in terms of the 2-Wasserstein distance and through its utility when used for downstream compression tasks.

Tingting Li, Ziming Zhao, Jianwei Yin

With the rapid advancements in quantum computing, cloud-based quantum services have gained increasing prominence. However, due to quantum noise, optimizing the deployment of quantum circuits remains an NP-hard problem with an expansive search space. Existing methods usually use heuristic algorithms to approximate the solution, such as the representative IBM Qiskit. On the one hand, they often find suboptimal deployment solutions. On the other hand, prior technologies do not consider user-specific requirements and can only provide a single deployment strategy. In this paper, we propose QCDeploy that can provide a ranked list of effective deployment strategies to optimize quantum serverless circuit deployment. Specifically, we model quantum circuits as Directed Acyclic Graph (DAG) representations and utilize graph contrastive learning for vector embedding. Then, a tailored list-aware learning-to-rank architecture is employed to generate a list of candidate strategies (prioritizing better strategies). We conduct extensive evaluations involving 45 prevalent quantum algorithm circuits across 3~5 qubits, utilizing 3 IBM quantum physical devices with three types of chip topologies. The results demonstrate that our proposed framework significantly outperforms IBMQ's default deployment scheme, e.g., achieving 17.95% overhead reduction and increasing the execution success rate by 20%~40%.

Ankan Dash, Jingyi Gu, Guiling Wang, Chen Chen

Virtual Reality (VR) headsets, while integral to the evolving digital ecosystem, present a critical challenge: the occlusion of users' eyes and portions of their faces, which hinders visual communication and may contribute to social isolation. To address this, we introduce RevAvatar, an innovative framework that leverages AI methodologies to enable reverse pass-through technology, fundamentally transforming VR headset design and interaction paradigms. RevAvatar integrates state-of-the-art generative models and multimodal AI techniques to reconstruct high-fidelity 2D facial images and generate accurate 3D head avatars from partially observed eye and lower-face regions. This framework represents a significant advancement in AI4Tech by enabling seamless interaction between virtual and physical environments, fostering immersive experiences such as VR meetings and social engagements. Additionally, we present VR-Face, a novel dataset comprising 200,000 samples designed to emulate diverse VR-specific conditions, including occlusions, lighting variations, and distortions. By addressing fundamental limitations in current VR systems, RevAvatar exemplifies the transformative synergy between AI and next-generation technologies, offering a robust platform for enhancing human connection and interaction in virtual environments.

Richard Levinson, Vinay Ravindra, Sreeja Roy-Singh

We present a novel Mixed Integer Linear Program formulation that produces optimal plans for a constellation of remote sensing satellites. The generalized formulation is applied to an operational NASA constellation to improve wildfire danger prediction. The planner generates integrated data collection and downlink plans for multiple agile satellites with limited storage capacity, minimum energy requirements, and temporal constraints. Observation targets and modes are associated with science rewards. The planner maximizes the aggregate rewards collected for all observations on all satellites. Our generalized model for integrated data collection and downlink uses a novel interval-based abstraction called Data Cycles, without time-indexed variables. Data cycles organize the multitude of observation and downlink opportunities from 1 second granularity into sequences of data collection and downlink intervals. Experiments using large-scale real-world data yield optimal 24-hr plans for an eight satellite constellation, which capture 99% of the ~23,000 available targets and 99.9% of available science rewards.

Jiaqi Zheng, Qing Ling, Yerong Feng

Although deep learning models have demonstrated remarkable potential in weather prediction, most of them overlook either the physics of the underlying weather evolution or the topology of the Earth’s surface. In light of these disadvantages, we develop PASSAT, a novel Physics-ASSisted And Topology-informed deep learning model for weather prediction. PASSAT attributes the weather evolution to two key factors: (i) the advection process that can be characterized by the advection equation and the Navier-Stokes equation; (ii) the Earth-atmosphere interaction that is difficult to both model and calculate. PASSAT also takes the topology of the Earth's surface into consideration, other than simply treating it as a plane. With these considerations, PASSAT numerically solves the advection equation and the Navier-Stokes equation on the spherical manifold, utilizes a spherical graph neural network to capture the Earth-atmosphere interaction, and generates the initial velocity fields that are critical to solving the advection equation from the same spherical graph neural network. In the 5.625-degree resolution ERA5 data set, PASSAT outperforms both the state-of-the-art deep learning-based weather prediction models and the operational numerical weather prediction model IFS T42.

Siyuan Yang, Cheng Song, Zhilu Lai, Wenjia Wang

Differential equations are involved in modeling many engineering problems. Many efforts have been devoted to solving differential equations. Due to the flexibility of neural networks, Physics Informed Neural Networks (PINNs) have recently been proposed to solve complex differential equations and have demonstrated superior performance in many applications. While the L2 loss function is usually a default choice in PINNs, it has been shown that the corresponding numerical solution is incorrect and unstable for some complex equations. In this work, we propose a new PINNs framework named Kernel Packet accelerated PINNs (KP-PINNs), which gives a new expression of the loss function using the reproducing kernel Hilbert space (RKHS) norm and uses the Kernel Packet (KP) method to accelerate the computation. Theoretical results show that KP-PINNs can be stable across various differential equations. Numerical experiments illustrate that KP-PINNs can solve differential equations effectively and efficiently. This framework provides a promising direction for improving the stability and accuracy of PINNs-based solvers in scientific computing.

Haoran Yang, Yinan Zhang, Qunshan He, Yuqi Ye, Jing Zhao, Wenhai Wang

As Industrial 4.0 unfolds and digital twin technology rapidly advances, modeling techniques that can abstract real-world industrial objects into accurate and robust models, referred to modeling for industrial objects (MIO) tasks, have become increasingly crucial. However, existing works still face two major limitations. First, each of these works primarily focuses on modeling a specific industrial object. When the industrial objects change, the proposed methods often struggle to adapt. Second, they fail to fully consider latent relationships within industrial data, limiting the model’s ability to leverage the data and resulting in suboptimal performance. To address these issues, we propose a novel modeling paradigm tailored for MIO tasks, named UltraModel. Specifically, a twin model graph module is designed to construct a customized graph based on the mechanisms of industrial objects and employ graph convolution to generate high-dimensional representations. Then, a multi-scale feature abstraction module and a spatial attention-based feature fusion module are proposed to complement each other in performing multi-scale feature abstraction and fusion on high-dimensional representations. Finally, the outputs are obtained by processing the fused representations through a feedforward network. Experiments on two different industrial objects demonstrate our UltraModel outperforms existing methods, offering a novel perspective for addressing industrial modeling challenges.

Hongxin Xiang, Jun Xia, Xin Jin, Wenjie Du, Li Zeng, Xiangxiang Zeng

Electron density (ED), which describes the probability distribution of electrons in space, is crucial for accurately understanding the energy and force distribution in molecular force fields (MFF). Existing machine learning force fields (MLFF) focus on mining appropriate physical quantities from the atom-level conformation to enhance the molecular geometry representation while ignoring the unique information from microscopic electrons. In this work, we propose an efficient Electronic Density representation framework to enhance molecular Geometric learning (called EDG), which leverages images rendered from ED to boost molecular geometric representations in MLFF. Specifically, we construct a novel image-based ED representation, which consists of 2 million 6-view images with RGB-D channels, and design an ED representation learning model, called ImageED, to learn ED-related knowledge from these images. We further propose an efficient ED-aware teacher and introduce a cross-modal distillation strategy to transfer knowledge from the image-based teacher to the geometry-based students. Extensive experiments on QM9 and rMD17 demonstrate that EDG can be directly integrated into existing geometry-based models and significantly improves the capabilities of these models (e.g., SchNet, EGNN, SphereNet, ViSNet) for geometry representation learning in MLFF with a maximum average performance increase of 33.7%. Code and appendix are available at https://github.com/HongxinXiang/EDG

Han Wan, Rui Zhang, Qi Wang, Yang Liu, Hao Sun

Accurately modeling and forecasting complex systems governed by partial differential equations (PDEs) is crucial in various scientific and engineering domains. However, traditional numerical methods struggle in real-world scenarios due to incomplete or unknown physical laws. Meanwhile, machine learning approaches often fail to generalize effectively when faced with scarce observational data and the challenge of capturing local and global features. To this end, we propose the Physics-encoded Spectral Attention Network (PeSANet), which integrates local and global information to forecast complex systems with limited data and incomplete physical priors. The model consists of two key components: a physics-encoded block that uses hard constraints to approximate local differential operators from limited data, and a spectral-enhanced block that captures long-range global dependencies in the frequency domain. Specifically, we introduce a novel spectral attention mechanism to model inter-spectrum relationships and learn long-range spatial features. Experimental results demonstrate that PeSANet outperforms existing methods across all metrics, particularly in long-term forecasting accuracy, providing a promising solution for simulating complex systems with limited data and incomplete physics.

Xiangxiang Shen, Zheng Wan, Lingfeng Wen, Licheng Sun, Jian Yang, Xuan Tang, Shing-Ho J. Lin, Xiao He, Mingsong Chen, Xian Wei

Crystal structures can be simplified as a periodic point set that repeats across three-dimensional space along an underlying lattice. Traditionally, crystal representation methods rely on descriptors such as lattice parameters, symmetry, and space groups to characterize the structure. However, in reality, atoms in materials always vibrate above absolute zero, causing their positions to fluctuate continuously. This dynamic behavior disrupts the fundamental periodicity of the lattice, making crystal graphs based on static lattice parameters and conventional descriptors discontinuous under slight perturbations. Chemists proposed the pairwise distance distribution (PDD) method to address this. However, the completeness of PDD requires defining a large number of neighboring atoms, leading to high computational costs. Additionally, PDD does not account for atomic information, making it challenging to apply it directly to crystal material property prediction tasks. To tackle these challenges, we introduce the atom-weighted Pairwise Distance Distribution (WPDD) and Unit cell Pairwise Distance Distribution (UPDD) for the first time, applying them to the construction of multi-edge crystal graphs. We demonstrate the continuity and general completeness of crystal graphs under slight atomic position perturbations. Moreover, by modeling PDD as global information and integrating it into matrix-based message passing, we significantly reduce computational costs. Comprehensive evaluation results show that WPDDFormer achieves state-of-the-art predictive accuracy across tasks on benchmark datasets such as the Materials Project and JARVIS-DFT.

Yichen Luo, Jia Wang, Dapeng Lan, Yu Liu, Zhibo Pang

Partial Differential Equations (PDEs) are fundamental for modeling physical systems, yet solving them in a generic and efficient manner using machine learning-based approaches remains challenging due to limited multi-input and multi-scale generalization capabilities, as well as high computational costs. This paper proposes the Multi-input and Multi-scale Efficient Transformer (MMET), a novel framework designed to address the above challenges. MMET decouples mesh and query points as two sequences and feeds them into the encoder and decoder, respectively, and uses a Gated Condition Embedding (GCE) layer to embed input variables or functions with varying dimensions, enabling effective solutions for multi-scale and multi-input problems. Additionally, a Hilbert curve-based reserialization and patch embedding mechanism decrease the input length. This significantly reduces the computational cost when dealing with large-scale geometric models. These innovations enable efficient representations and support multi-scale resolution queries for large-scale and multi-input PDE problems. Experimental evaluations on diverse benchmarks spanning different physical fields demonstrate that MMET outperforms SOTA methods in both accuracy and computational efficiency. This work highlights the potential of MMET as a robust and scalable solution for real-time PDE solving in engineering and physics-based applications, paving the way for future explorations into pre-trained large-scale models in specific domains. This work is open-sourced at https://github.com/YichenLuo-0/MMET.

Jiaqi Liu, Jiaxu Cui, Shiang Sun, Yizhu Zhao, Bo Yang

With advancements in deep learning, neural simulators have become increasingly important for improving the efficiency and effectiveness of simulating complex dynamical systems in various scientific and technological fields. This paper presents a novel neural simulator called Context-informed Polymorphic Neural ODE Processes (CoPoNDP), aimed at addressing the challenges of modeling dynamical systems encountering concurrent environmental and temporal distribution shifts, which are common in real-world scenarios. CoPoNDP employs a context-driven neural stochastic process governed by a combination of basic differential equations in a time-sensitive manner to adaptively modulate the evolution of system states. This allows for flexible adaptation to changing temporal dynamics and generalization across different environments. Extensive experiments conducted on dynamical systems from ecology, chemistry, physics, and energy demonstrate that by effectively utilizing contextual information, CoPoNDP outperforms the state-of-the-art models in handling joint distribution shifts. It also shows robustness in sparse and noisy settings, making it a promising approach for modeling dynamical systems in complex real-world applications.

Xinjie Li, Abhinav Verma

Single-step retrosynthesis is a crucial task in organic synthesis, where the objective is to identify the reactants needed to produce a given product. In recent years, a variety of machine learning methods have been developed to tackle retrosynthesis prediction. In our study, we introduce RetroMoE, a novel generative model designed for the single-step retrosynthesis task. We start with a non-symmetric variational autoencoder (VAE) that incorporates a graph encoder to map molecular graphs into a latent space, followed by a transformer decoder for precise prediction of molecular SMILES strings. Additionally, we implement a simple yet effective mixture-of-experts (MoE) network to translate the product latent embedding into the reactant latent embedding. To our knowledge, this is the first approach that frames single-step retrosynthesis as a latent translation problem. Extensive experiments on the USPTO-50K and USPTO-MIT datasets demonstrate the superiority of our method, which not only surpasses most semi-template-based and template-free methods but also delivers competitive results against template-based methods. Notably, under the class-known setting on the USPTO-50K, our method achieves top-1 exact match accuracy comparable to the state-of-the-art template method, RetroKNN.

Tianyu Chen, Haoyi Zhou, Ying Li, Hao Wang, Zhenzhe Zhang, Tianchen Zhu, Shanghang Zhang, Jianxin Li

Fourier Neural Operators (FNO) have emerged as promising solutions for efficiently solving partial differential equations (PDEs) by learning infinite-dimensional function mappings through frequency domain transformations. However, the sparsity of high-frequency signals limits computational efficiency for high-dimensional inputs, and fixed-pattern truncation often causes high-frequency signal loss, reducing performance in scenarios such as high-resolution inputs or long-term predictions. To address these challenges, we propose FreqMoE, an efficient and progressive training framework that exploits the dependency of high-frequency signals on low-frequency components. The model first learns low-frequency weights and then applies a sparse upward-cycling strategy to construct a mixture of experts (MoE) in the frequency domain, effectively extending the learned weights to high-frequency regions. Experiments on both regular and irregular grid PDEs demonstrate that FreqMoE achieves up to 16.6 percent accuracy improvement while using merely 2.1 percent parameters (47.32x reduction) compared to dense FNO. Furthermore, the approach demonstrates remarkable stability in long-term predictions and generalizes seamlessly to various FNO variants and grid structures, establishing a new Low frequency Pretraining, High frequency Fine-tuning'' paradigm for solving PDEs.

Chenhong Zhou, Jie Chen, Zaifeng Yang, Ching Eng Png

Physics-informed neural networks (PINNs) have emerged as a new learning paradigm for solving partial differential equations (PDEs) by enforcing the constraints of physical equations, boundary conditions (BCs), and initial conditions (ICs) into the loss function. Despite their successes, vanilla PINNs still suffer from poor accuracy and slow convergence due to the intractable multi-objective optimization issue. In this paper, we propose a novel Dual-Balanced PINN (DB-PINN), which dynamically adjusts loss weights by integrating inter-balancing and intra-balancing to alleviate two imbalance issues in PINNs. Inter-balancing aims to mitigate the gradient imbalance between PDE residual loss and condition-fitting losses by determining an aggregated weight that offsets their gradient distribution discrepancies. Intra-balancing acts on condition-fitting losses to tackle the imbalance in fitting difficulty across diverse conditions. By evaluating the fitting difficulty based on the loss records, intra-balancing can allocate the aggregated weight proportionally to each condition loss according to its fitting difficulty level. We further introduce a robust weight update strategy to prevent abrupt spikes and arithmetic overflow in instantaneous weight values caused by large loss variances, enabling smooth weight updating and stable training. Extensive experiments demonstrate that DB-PINN achieves significantly superior performance than those popular gradient-based weighting methods in terms of convergence speed and prediction accuracy. Our code and supplementary material are available at https://github.com/chenhong-zhou/DualBalanced-PINNs.

Chong Yu, Tao Chen, Zhongxue Gan

The rapid improvement of deep learning models with the integration of the physical world has dramatically improved embodied AI capabilities. Meanwhile, the powerful embodied AI models and their scales place an increasing burden on deployment efficiency. The efficiency issue is more apparent on embodied AI platforms than on data centers because they have more limited computational resources and memory bandwidth. Meanwhile, most embodied AI scenarios, like autonomous driving and robotics, are more sensitive to fast responses. Theoretically, the traditional model compression techniques can help embodied AI models with more efficient computation, lower memory and energy consumption, and reduced latency. Because the embodied AI models are expected to interact with the physical world, the corresponding compressed models are also expected to resist natural corruption caused by real-world events such as noise, blur, weather conditions, and even adversarial corruption. This paper explores the novel paradigm to boost the efficiency of the embodied AI models and the robust compression boundary. The efficacy of our method has been proven to find the optimal balance between accuracy, efficiency, and robustness in real-world conditions.

Xianliang Xu, Ye Li, Zhongyi Huang

In recent years, neural networks have achieved remarkable progress in various fields and have also drawn much attention in applying them on scientific problems. A line of methods involving neural networks for solving partial differential equations (PDEs), such as Physics-Informed Neural Networks (PINNs) and the Deep Ritz Method (DRM), has emerged. Although these methods outperform classical numerical methods in certain cases, the optimization problems involving neural networks are typically non-convex and non-smooth, which can result in unsatisfactory solutions for PDEs. In contrast to deterministic neural networks, the hidden weights of random neural networks are sampled from some prior distribution and only the output weights participate in training. This makes training much simpler, but it remains unclear how to select the prior distribution. In this paper, we focus on Barron type functions and approximate them under Sobolev norms by random neural networks with clear prior distribution. In addition to the approximation error, we also derive bounds for the optimization and generalization errors of random neural networks for solving PDEs when the solutions are Barron type functions.