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
论文检索
输入标题、作者或关键词，从 1,767 篇学术成果中精准定位
Towards Generalizable Neural Simulators: Addressing Distribution Shifts Induced by Environmental and Temporal Variations
PDF ↗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.
Existing graph neural network (GNN) methods are typically built upon the i.i.d. assumption, emphasizing the enhancement of the test performance for in-distribution (ID) data. However, there has been limited exploration of their adaptability to scenarios involving unknown distribution data. On the one hand, in real-world application scenarios, graph data often expands continuously with the acquisition of external knowledge, which means that new nodes with unknown categories may be added to the graph data. The gap between the new node distribution and the original node distribution can make existing GNN methods less effective. On the other hand, existing out-of-distribution (OOD) detection methods often rely on the softmax confidence score, which makes the OOD data suffer from overconfident posterior distributions. To address the above issues, we propose an Energy Propagation-based Graph Neural Network (EPGNN), which improves the OOD generalization ability by endowing GNN with the capacity to detect the OOD nodes in the graph. Specifically, we first construct GNN encoder to obtain node embedding that incorporates neighborhood structural information. Then, we design a plug-and-play energy-based OOD evaluator by assigning corresponding energy values to different nodes. Finally, we construct a plug-and-play structure-aware energy propagation module and joint alignment regularization, which make the node energy more flexible during the training process. Extensive experiments on benchmark datasets demonstrate the superiority of our method.
Most edge-cloud collaboration frameworks rely on the substantial computational and storage capabilities of cloud-based artificial neural networks (ANNs). However, this reliance results in significant communication overhead between edge devices and the cloud, as well as high computational energy consumption, especially when applied to resource-constrained edge devices. To address these challenges, we propose ECC-SNN, a novel edge-cloud collaboration framework that incorporates energy-efficient spiking neural networks (SNNs) to offload more computational workload from the cloud to the edge, thereby improving cost-effectiveness and reducing reliance on the cloud. ECC-SNN employs a joint training approach that integrates ANN and SNN models, enabling edge devices to leverage knowledge from cloud models for enhanced performance while reducing energy consumption and processing latency. Furthermore, ECC-SNN features an on-device incremental learning algorithm that enables edge models to continuously adapt to dynamic environments, reducing the communication overhead and resource consumption associated with frequent cloud update requests. Extensive experimental results on four datasets demonstrate that ECC-SNN improves accuracy by 4.15%, reduces average energy consumption by 79.4%, and lowers average processing latency by 39.1%.
The energy efficiency of deep spiking neural networks (SNNs) aligns with the constraints of resource-limited edge devices, positioning SNNs as a promising foundation for intelligent applications leveraging the extensive data collected by these devices. To safeguard data privacy, federated learning (FL) facilitates collaborative SNN-based model training by leveraging data distributed across edge devices without transmitting local data to a central server. However, existing FL approaches encounter challenges in handling label-skewed data across devices, inducing drift in the local SNN model and consequently impairing the performance of the global SNN model. To tackle these problems, we propose a novel framework called FedLEC, which incorporates intra-client label weight calibration to balance the learning intensity across local labels and inter-client knowledge distillation to mitigate local SNN model bias caused by label absence. Extensive experiments with three different structured SNNs across five datasets (i.e., three non-neuromorphic and two neuromorphic datasets) demonstrate the efficiency of FedLEC. Compared to seven state-of-the-art FL algorithms, FedLEC achieves an average accuracy improvement of approximately 11.59% for the global SNN model under various label skew distribution settings.
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.
Over-smoothing is a persistent challenge in Graph Neural Networks (GNNs), where node embeddings become indistinguishable as network depth increases, fundamentally limiting their effectiveness on tasks requiring fine-grained distinctions. This issue arises from the reliance on diffusion-based propagation mechanisms, which suppress high-frequency information essential for preserving feature diversity. To mitigate this, we propose a wave-driven GNN framework that redefines feature propagation through the wave equation. Unlike diffusion, the wave equation incorporates second-order dynamics, balancing smoothing with oscillatory behavior to retain high-frequency components while ensuring effective information flow. To enhance the stability and convergence of wave equation discretization on graphs, an energy-based mechanism inspired by kinetic and potential energy dynamics is introduced, balancing temporal evolution and structural alignment to stabilize propagation. Extensive experiments on benchmark datasets, including Cora, Citeseer, and PubMed, as well as real-world graphs, demonstrate that the proposed framework achieves state-of-the-art performance, effectively mitigating over-smoothing and enabling deeper, more expressive architectures. The code is available at https://github.com/rene0329/EWGNN/.
Heterophilic graph neural networks (GNNs) have gained prominence for their ability to learn effective representations in graphs with diverse, attribute-aware relationships. While existing methods leverage attribute inference during message passing to improve performance, they often struggle with challenging heterophilic graphs. This is due to edge distribution shifts introduced by diverse connection patterns, which blur attribute distinctions and undermine message-passing stability. This paper introduces H₂OGNN, a novel framework that reframes edge attribute inference as an out-of-distribution (OOD) detection problem. H₂OGNN introduces a simple yet effective symbolic energy regularization approach for OOD learning, ensuring robust classification boundaries between homophilic and heterophilic edge attributes. This design significantly improves the stability and reliability of GNNs across diverse connectivity patterns. Through theoretical analysis, we show that H₂OGNN addresses the graph denoising problem by going beyond feature smoothing, offering deeper insights into how precise edge attribute identification boosts model performance. Extensive experiments on nine benchmark datasets demonstrate that H₂OGNN not only achieves state-of-the-art performance but also consistently outperforms other heterophilic GNN frameworks, particularly on datasets with high heterophily.
QuantileFormer: Probabilistic Time Series Forecasting with a Pattern-Mixture Decomposed VAE Transformer
PDF ↗Probabilistic time series forecasting has attracted an increasing attention in machine learning community for its potential applications in the fields of renewable energy, traffic management, healthcare, etc. Previous research mainly focused on extracting long-range dependencies for point-wise prediction, which fail to capture complex temporal patterns and statistical characteristics for probabilistic analysis. In this paper, we propose a novel pattern-mixture decomposition method that decomposes long-term series into quantile drift, divergence patterns, and Gaussian mixture components, which can effectively capture the intricate temporal patterns and stochastic characteristics in time series. Based on pattern-mixture decomposition, we propose a novel Transformer-based model called QuantileFormer for probabilistic time series forecasting. It takes the the comprehensive drift-divergence mixture patterns as features, and designs a variational inference based fusion Transformer architecture to generate quantile prediction results. Extensive experiments show that the proposed method consistently boosts the baseline methods by a large margin and achieves state-of-the-art performance on six real-world benchmarks.
The combination of Spiking Neural Networks (SNNs) with Vision Transformer architectures has attracted significant attention due to the great potential for energy-efficient and high-performance computing paradigms. However, a substantial performance gap still exists between SNN-based and ANN-based transformer architectures. While existing methods propose spiking self-attention mechanisms that are successfully combined with SNNs, the overall architectures proposed by these methods suffer from a bottleneck in effectively extracting features from different image scales. In this paper, we address this issue and propose MSVIT, a novel spike-driven Transformer architecture, which firstly uses multi-scale spiking attention (MSSA) to enrich the capability of spiking attention blocks. We validate our approach across various main data sets. The experimental results indicate that our MSVIT outperforms existing SNN-based models, positioning itself as a state-of-the-art solution among NN-transformer architectures. The codes are available at https://github.com/Nanhu-AI-Lab/MSViT.
Temporal processing is vital for extracting meaningful information from time-varying signals. Recent advancements in Spiking Neural Networks (SNNs) have shown immense promise in efficiently processing these signals. However, progress in this field has been impeded by the lack of effective and standardized benchmarks, which complicates the consistent measurement of technological advancements and limits the practical applicability of SNNs. To bridge this gap, we introduce the Neuromorphic Sequential Arena (NSA), a comprehensive benchmark that offers an effective, versatile, and application-oriented evaluation framework for neuromorphic temporal processing. The NSA includes seven real-world temporal processing tasks from a diverse range of application scenarios, each capturing rich temporal dynamics across multiple timescales. Utilizing NSA, we conduct extensive comparisons of recently introduced spiking neuron models and neural architectures, presenting comprehensive baselines in terms of task performance, training speed, memory usage, and energy efficiency. Our findings emphasize an urgent need for efficient SNN designs that can consistently deliver high performance across tasks with varying temporal complexities while maintaining low computational costs. NSA enables systematic tracking of advancements in neuromorphic algorithm research and paves the way for developing effective and efficient neuromorphic temporal processing systems.
Case-based prediction (CBP) methods do not learn a model of the target decision function but instead perform an inference process that depends on two similarity measures and a reference case base. This paper proposes a strategy, called EnergyCompress, to learn an effective case base by selecting relevant cases from an initial set. Use of EnergyCompress decreases CBP inference time, through case base compression, and also increases prediction performance, for a wide variety of CBP algorithms. EnergyCompress relies on the proposition of a general formulation of the CBP task in the framework of energy-based models, which leads to a new and valuable characterization of the notion of competence in case-based reasoning, in particular at the source case level. Extensive experimental results on 18 benchmark datasets comparing EnergyCompress to 5 reference algorithms for case base maintenance support the benefit of the proposed strategy.
ILIF: Temporal Inhibitory Leaky Integrate-and-Fire Neuron for Overactivation in Spiking Neural Networks
PDF ↗The Spiking Neural Network (SNN) has drawn increasing attention for its energy-efficient, event-driven processing and biological plausibility. To train SNNs via backpropagation, surrogate gradients are used to approximate the non-differentiable spike function, but they only maintain nonzero derivatives within a narrow range of membrane potentials near the firing threshold—referred to as the surrogate gradient support width gamma. We identify a major challenge, termed the dilemma of gamma: a relatively large gamma leads to overactivation, characterized by excessive neuron firing, which in turn increases energy consumption, whereas a small gamma causes vanishing gradients and weakens temporal dependencies. To address this, we propose a temporal Inhibitory Leaky Integrate-and-Fire (ILIF) neuron model, inspired by biological inhibitory mechanisms. This model incorporates interconnected inhibitory units for membrane potential and current, effectively mitigating overactivation while preserving gradient propagation. Theoretical analysis demonstrates ILIF’s effectiveness in overcoming the gamma dilemma, and extensive experiments on multiple datasets show that ILIF improves energy efficiency by reducing firing rates, stabilizes training, and enhances accuracy. The code is available at github.com/kaisun1/ILIF.
Transformer-based Spiking Neural Networks (SNNs) introduce a novel event-driven self-attention paradigm that combines the high performance of Transformers with the energy efficiency of SNNs. However, the larger model size and increased computational demands of the Transformer structure limit their practicality in resource-constrained scenarios. In this paper, we integrate binarization techniques into Transformer-based SNNs and propose the Binary Event-Driven Spiking Transformer, i.e. BESTformer. The proposed BESTformer can significantly reduce storage and computational demands by representing weights and attention maps with a mere 1-bit. However, BESTformer suffers from a severe performance drop from its full-precision counterpart due to the limited representation capability of binarization. To address this issue, we propose a Coupled Information Enhancement (CIE) method, which consists of a reversible framework and information enhancement distillation. By maximizing the mutual information between the binary model and its full-precision counterpart, the CIE method effectively mitigates the performance degradation of the BESTformer. Extensive experiments on static and neuromorphic datasets demonstrate that our method achieves superior performance to other binary SNNs, showcasing its potential as a compact yet high-performance model for resource-limited edge devices. The repository of this paper is available at https://github.com/CaoHLin/BESTFormer.
This paper introduces a novel mechanism for online allocation with multi-phase, non-separable regularizers, termed Cap-and-Penalize (CnP), inspired by real-world applications such as cap-and-tax policies in carbon pricing. The CnP regularizer models a multi-phase cost structure, imposing a monotone convex penalty when total allocation exceeds a predefined level (soft cap) and enforcing a strict limit (hard cap) beyond which allocation is prohibited. Our contributions are twofold: (1) we propose an online mechanism for CnP-regularized allocation without per-step resource constraints, which operates as a simple and intuitive posted-price mechanism, but achieves the best-possible guarantee among all possible online algorithms; (2) we tackle the more complex setting with per-step resource constraints by decomposing the regularizer into local components, yielding a similar mechanism with time-dependent marginal pricing functions. To establish the tightness of our results in both settings, we introduce a representative function-based approach that transforms the lower-bound proof into the problem of solving an ordinary differential equation with boundary conditions. We believe that this technique has the potential to be applied to other similar online optimization problems.
On-device sequential recommendation (SR) systems are designed to make local inferences using real-time features, thereby alleviating the communication burden on server-based recommenders when handling concurrent requests from millions of users. However, the resource constraints of edge devices, including limited memory and computational capacity, pose significant challenges to deploying efficient SR models. Inspired by the energy-efficient and sparse computing properties of deep Spiking Neural Networks (SNNs), we propose a cost-effective on-device SR model named SSR, which encodes dense embedding representations into sparse spike-wise representations and integrates novel spiking filter modules to extract temporal patterns and critical features from item sequences, optimizing computational and memory efficiency without sacrificing recommendation accuracy. Extensive experiments on real-world datasets demonstrate the superiority of SSR. Compared to other SR baselines, SSR achieves comparable recommendation performance while reducing energy consumption by an average of 59.43%. In addition, SSR significantly lowers memory usage, making it particularly well-suited for deployment on resource-constrained edge devices.
RePST: Language Model Empowered Spatio-Temporal Forecasting via Semantic-Oriented Reprogramming
PDF ↗Spatio-temporal forecasting is pivotal in numerous real-world applications, including transportation planning, energy management, and climate monitoring. In this work, we aim to harness the reasoning and generalization abilities of Pre-trained Language Models (PLMs) for more effective spatio-temporal forecasting, particularly in data-scarce scenarios. However, recent studies uncover that PLMs, which are primarily trained on textual data, often falter when tasked with modeling the intricate correlations in numerical time series, thereby limiting their effectiveness in comprehending spatio-temporal data. To bridge the gap, we propose RePST, a semantic-oriented PLM reprogramming framework tailored for spatio-temporal forecasting. Specifically, we first propose a semantic-oriented decomposer that adaptively disentangles spatially correlated time series into interpretable sub-components, which facilitates PLM to understand sophisticated spatio-temporal dynamics via a divide-and-conquer strategy. Moreover, we propose a selective discrete reprogramming scheme, which introduces an expanded spatio-temporal vocabulary space to project spatio-temporal series into discrete representations. This scheme minimizes the information loss during reprogramming and enriches the representations derived by PLMs. Extensive experiments on real-world datasets show that the proposed RePST outperforms twelve state-of-the-art baseline methods, particularly in data-scarce scenarios, highlighting the effectiveness and superior generalization capabilities of PLMs for spatio-temporal forecasting. Codes and Appendix can be found at https://github.com/usail-hkust/REPST.
Foundation models, such as CNNs and ViTs, have powered the development of image representation learning. However, general guidance to model architecture design is still missing. Inspired by the connection between image representation learning and heat conduction, we model images by the heat conduction equation, where the essential idea is to conceptualize image features as temperatures and model their information interaction as the diffusion of thermal energy. Based on this idea, we find that many modern model architectures, such as residual structures, SE block, and feed-forward networks, can be interpreted from the perspective of the heat conduction equation. Therefore, we leverage the heat equation to design new and more interpretable models. As an example, we propose the Heat Conduction Layer and the Refinement Approximation Layer inspired by solving the heat conduction equation using Finite Difference Method and Fourier series, respectively. The main goal of this paper is to integrate the overall architectural design of neural networks into the theoretical framework of heat conduction. Nevertheless, our Heat Conduction Network (HcNet) still shows competitive performance, e.g., HcNet-T achieves 83.0% top-1 accuracy on ImageNet-1K while only requiring 28M parameters and 4.1G MACs. The code is publicly available at: https://github.com/ZheminZhang1/HcNet.
This paper tackles the challenge of anomaly image synthesis and segmentation to generate various anomaly images and their segmentation labels to mitigate the issue of data scarcity. Existing approaches employ the precise mask to guide the generation, relying on additional mask generators, leading to increased computational costs and limited anomaly diversity. Although a few works use coarse masks as the guidance to expand diversity, they lack effective generation of labels for synthetic images, thereby reducing their practicality. Therefore, our proposed method simultaneously generates anomaly images and their corresponding masks by utilizing coarse masks and anomaly categories. The framework utilizes attention maps from synthesis process as mask labels and employs two optimization modules to tackle drift challenges, which are mismatches between synthetic results and real situations. Our evaluation demonstrates that our method improves pixel-level AP by 1.3% and F1-MAX by 1.8% in anomaly detection tasks on the MVTec dataset. Additionally, its successful application in practical scenarios highlights its effectiveness, improving IoU by 37.2% and F-measure by 25.1% with the Floor Dirt dataset. The code is available at https://github.com/JJessicaYao/DriftRemover.
Economic growth and development require a consistent supply of energy. Energy which has mainly been supplied from fossil fuels. The impacts of these on the environment such as global warming, raised an alarm on their use. As a result other sources of energy such as wind energy are used as alternatives for electricity production. Wind energy assessment nevertheless faces barriers due to its stochastic nature. This later creates various regimes, which traditional models can't always fit thereby producing poor estimates. In this work, we aim to use Large Language Models (LLMs) to predict the wind potential in a given location. Through this approach, we aim at lifting the barrier on energy problems in developing countries by providing knowledge on the state of wind energy in given locations.