Wei Gong

Luoyu Mei, Shuai Wang, Yun Cheng et al.

Semantic recognition is pivotal in virtual reality (VR) applications, enabling immersive and interactive experiences. A promising approach is utilizing millimeter-wave (mmWave) signals to generate point clouds. However, the high computational and memory demands of current mmWave point cloud models hinder their efficiency and reliability. To address this limitation, our paper introduces ESP-PCT, a novel Enhanced Semantic Performance Point Cloud Transformer with a two-stage semantic recognition framework tailored for VR applications. ESP-PCT takes advantage of the accuracy of sensory point cloud data and optimizes the semantic recognition process, where the localization and focus stages are trained jointly in an end-to-end manner. We evaluate ESP-PCT on various VR semantic recognition conditions, demonstrating substantial enhancements in recognition efficiency. Notably, ESP-PCT achieves a remarkable accuracy of 93.2% while reducing the computational requirements (FLOPs) by 76.9% and memory usage by 78.2% compared to the existing Point Transformer model simultaneously. These underscore ESP-PCT's potential in VR semantic recognition by achieving high accuracy and reducing redundancy. The code and data of this project are available at \url{https://github.com/lymei-SEU/ESP-PCT}.

Wei Gong, Hehu Xie, Ningning Yan

We propose in this paper a multilevel correction method to solve optimal control problems constrained by elliptic equations with the finite element method. In this scheme, solving optimization problem on the finest finite element space is transformed to a series of solutions of linear boundary value problems by the multigrid method on multilevel meshes and a series of solutions of optimization problems on the coarsest finite element space. Our proposed scheme, instead of solving a large scale optimization problem in the finest finite element space, solves only a series of linear boundary value problems and the optimization problems in a very low dimensional finite element space, and thus can improve the overall efficiency for the solution of optimal control problems governed by PDEs.

Wei Gong, Ningning Yan, Zhaojie Zhou

This paper aims to study the convergence of adaptive finite element method for control constrained elliptic optimal control problems under $L^2$-norm. We prove the contraction property and quasi-optimal complexity for the $L^2$-norm errors of both the control, the state and adjoint state variables with $L^2$-norm based AFEM, this is in contrast to and improve our previous work [13] where convergence of AFEM based on energy norm had been studied and suboptimal convergence for the control variable was obtained and observed numerically. For the discretization we use variational discretization for the control and piecewise linear and continuous finite elements for the state and adjoint state. Under mild assumptions on the initial mesh and the mesh refinement algorithm to keep the adaptive meshes sufficiently mildly graded we prove the optimal convergence of AFEM for the control problems, numerical results are provided to support our theoretical findings.

Echo Diyun LU, Charles Findling, Marianne Clausel et al.

Regime transitions routinely break stationarity in time series, making calibrated uncertainty as important as point accuracy. We study distribution-free uncertainty for regime-switching forecasting by coupling Deep Switching State Space Models with Adaptive Conformal Inference (ACI) and its aggregated variant (AgACI). We also introduce a unified conformal wrapper that sits atop strong sequence baselines including S4, MC-Dropout GRU, sparse Gaussian processes, and a change-point local model to produce online predictive bands with finite-sample marginal guarantees under nonstationarity and model misspecification. Across synthetic and real datasets, conformalized forecasters achieve near-nominal coverage with competitive accuracy and generally improved band efficiency.

Linxiao Cao, Yifei Zhu, Wei Gong

Large pre-trained models have exhibited remarkable achievements across various domains. The substantial training costs associated with these models have led to wide studies of fine-tuning for effectively harnessing their capabilities in solving downstream tasks. Yet, conventional fine-tuning approaches become infeasible when the model lacks access to downstream data due to privacy concerns. Naively integrating fine-tuning approaches with the emerging federated learning frameworks incurs substantial communication overhead and exerts high demand on local computing resources, making it impractical for common resource-limited devices. In this paper, we introduce SFPrompt, an innovative privacy-preserving fine-tuning method tailored for the federated setting where direct uploading of raw data is prohibited and local devices are resource-constrained to run a complete pre-trained model. In essence, SFPrompt judiciously combines split learning with federated learning to handle these challenges. Specifically, the pre-trained model is first partitioned into client and server components, thereby streamlining the client-side model and substantially alleviating computational demands on local resources. SFPrompt then introduces soft prompts into the federated model to enhance the fine-tuning performance. To further reduce communication costs, a novel dataset pruning algorithm and a local-loss update strategy are devised during the fine-tuning process. Extensive experiments demonstrate that SFPrompt delivers competitive performance as the federated full fine-tuning approach while consuming a mere 0.46% of local computing resources and incurring 53% less communication cost.

Wei Gong, Chuwen Ma, Ziyi Zhang

In this paper, we propose an unfitted finite element method to solve PDE-constrained shape optimization problems via shape gradient flow. The shape gradient flow system consists of the state equation, the adjoint equation, the velocity equation, as well as the flow map that generates the evolution of the boundary driven by the velocity field, which can be viewed as a limit system of the classical shape gradient descent algorithm. In \cite{GongLiRao} the authors proposed an evolving finite element method to solve the shape gradient flow system. Instead, in this paper, we propose an unfitted finite element method in which the evolution of the boundary is realized by cubic splines and the equations are solved by cut finite element methods with ghost penalization. Under reasonable assumptions, we are able to prove some optimal convergence rates that are further validated by numerical experiments.

Yunyun Feng, Chenhong Cao, Si Chen et al.

5G backscatter communication presents an emerging energy-efficient IoT connectivity solution with enhanced availability and data rate advantages over traditional wireless networks. For 5G backscatter, synchronization is crucial as it ensures high-quality transmission. Popular synchronization methods employ autocorrelation and cross-correlation for accurate timing, yet they are constrained by resources. Traditional cross-correlation-based methods for resource utilization optimization also fail in 5G backscatter due to the presence of multiple templates for 5G. A synchronization strategy that supports high accuracy and low power would be highly attractive for wireless backscatter communication. We propose Symmetric Differential (SD)-based Sync, an accurate and resource-efficient synchronization method for 5G backscatter. We have observed that the envelope of the 5G Primary Synchronization Signal (PSS) exhibits a unique mirror symmetry, which enables us to employ differential techniques for low-power PSS detection. We extensively evaluated our design using a testbed of backscatter hardware, SDR gNodeB, and User Equipment (UE). Results show that our SD consumes 3,175 D flip-flops, which is 87x lower than NR fine timing (NFT), 181x lower than symmetry-based semi-template sync (SST), and 30x lower than symmetric autocorrelation (SA)-based sync.

Fan Xu, Hao Wu, Nan Wang et al.

The modeling of complicated time-evolving physical dynamics from partial observations is a long-standing challenge. Particularly, observations can be sparsely distributed in a seemingly random or unstructured manner, making it difficult to capture highly nonlinear features in a variety of scientific and engineering problems. However, existing data-driven approaches are often constrained by fixed spatial and temporal discretization. While some researchers attempt to achieve spatio-temporal continuity by designing novel strategies, they either overly rely on traditional numerical methods or fail to truly overcome the limitations imposed by discretization. To address these, we propose CoPS, a purely data-driven methods, to effectively model continuous physics simulation from partial observations. Specifically, we employ multiplicative filter network to fuse and encode spatial information with the corresponding observations. Then we customize geometric grids and use message-passing mechanism to map features from original spatial domain to the customized grids. Subsequently, CoPS models continuous-time dynamics by designing multi-scale graph ODEs, while introducing a Markov-based neural auto-correction module to assist and constrain the continuous extrapolations. Comprehensive experiments demonstrate that CoPS advances the state-of-the-art methods in space-time continuous modeling across various scenarios.

Binchuan Qi, Wei Gong, Li Li

In this paper, we adopt a probability distribution estimation perspective to explore the optimization mechanisms of supervised classification using deep neural networks. We demonstrate that, when employing the Fenchel-Young loss, despite the non-convex nature of the fitting error with respect to the model's parameters, global optimal solutions can be approximated by simultaneously minimizing both the gradient norm and the structural error. The former can be controlled through gradient descent algorithms. For the latter, we prove that it can be managed by increasing the number of parameters and ensuring parameter independence, thereby providing theoretical insights into mechanisms such as over-parameterization and random initialization. Ultimately, the paper validates the key conclusions of the proposed method through empirical results, illustrating its practical effectiveness.

Yuhan Pan, Yanan Sun, Wei Gong

Long-Tailed (LT) recognition has been widely studied to tackle the challenge of imbalanced data distributions in real-world applications. However, the design of neural architectures for LT settings has received limited attention, despite evidence showing that architecture choices can substantially affect performance. This paper aims to bridge the gap between LT challenges and neural network design by providing an in-depth analysis of how various architectures influence LT performance. Specifically, we systematically examine the effects of key network components on LT handling, such as topology, convolutions, and activation functions. Based on these observations, we propose two convolutional operations optimized for improved performance. Recognizing that operation interactions are also crucial to network effectiveness, we apply Neural Architecture Search (NAS) to facilitate efficient exploration. We propose LT-DARTS, a NAS method with a novel search space and search strategy specifically designed for LT data. Experimental results demonstrate that our approach consistently outperforms existing architectures across multiple LT datasets, achieving parameter-efficient, state-of-the-art results when integrated with current LT methods.

Fan Xu, Wei Gong, Hao Wu et al.

Wildfires, as an integral component of the Earth system, are governed by a complex interplay of atmospheric, oceanic, and terrestrial processes spanning a vast range of spatiotemporal scales. Modeling their global activity on large timescales is therefore a critical yet challenging task. While deep learning has recently achieved significant breakthroughs in global weather forecasting, its potential for global wildfire behavior prediction remains underexplored. In this work, we reframe this problem and introduce the Hierarchical Graph ODE (HiGO), a novel framework designed to learn the multi-scale, continuous-time dynamics of wildfires. Specifically, we represent the Earth system as a multi-level graph hierarchy and propose an adaptive filtering message passing mechanism for both intra- and inter-level information flow, enabling more effective feature extraction and fusion. Furthermore, we incorporate GNN-parameterized Neural ODE modules at multiple levels to explicitly learn the continuous dynamics inherent to each scale. Through extensive experiments on the SeasFire Cube dataset, we demonstrate that HiGO significantly outperforms state-of-the-art baselines on long-range wildfire forecasting. Moreover, its continuous-time predictions exhibit strong observational consistency, highlighting its potential for real-world applications.

Fan Xu, Hao Wu, Kun Wang et al.

In dynamical system modeling, traditional numerical methods are limited by high computational costs, while modern data-driven approaches struggle with data scarcity and distribution shifts. To address these fundamental limitations, we first propose SPARK, a physics-guided quantitative augmentation plugin. Specifically, SPARK utilizes a reconstruction autoencoder to integrate physical parameters into a physics-rich discrete state dictionary. This state dictionary then acts as a structured dictionary of physical states, enabling the creation of new, physically-plausible training samples via principled interpolation in the latent space. Further, for downstream prediction, these augmented representations are seamlessly integrated with a Fourier-enhanced Graph ODE, a combination designed to robustly model the enriched data distribution while capturing long-term temporal dependencies. Extensive experiments on diverse benchmarks demonstrate that SPARK significantly outperforms state-of-the-art baselines, particularly in challenging out-of-distribution scenarios and data-scarce regimes, proving the efficacy of our physics-guided augmentation paradigm.

Zhipeng Zhou, Liu Liu, Peilin Zhao et al.

Multi-task learning (MTL) has emerged as a promising approach for deploying deep learning models in real-life applications. Recent studies have proposed optimization-based learning paradigms to establish task-shared representations in MTL. However, our paper empirically argues that these studies, specifically gradient-based ones, primarily emphasize the conflict issue while neglecting the potentially more significant impact of imbalance/dominance in MTL. In line with this perspective, we enhance the existing baseline method by injecting imbalance-sensitivity through the imposition of constraints on the projected norms. To demonstrate the effectiveness of our proposed IMbalance-sensitive Gradient (IMGrad) descent method, we evaluate it on multiple mainstream MTL benchmarks, encompassing supervised learning tasks as well as reinforcement learning. The experimental results consistently demonstrate competitive performance.

Binchuan Qi, Wei Gong, Li Li

Despite its empirical success, deep learning still lacks a comprehensive theoretical understanding of model fitting and generalization. This paper proposes the probability distribution (PD) learning framework to analyze the optimization and generalization mechanisms of deep learning. Within this framework, the conditional distribution of labels given features is the primary learning target, with the loss function, prior knowledge, and model properties explicitly characterized. Under these formulations, we establish theoretical guarantees on optimizability, even in non-convex settings, and derive generalization error bounds that provide meaningful explanations for practical performance. Specifically, we first prove theoretically that the Fenchel-Young loss is the natural and necessary choice for solving PD learning problems, thereby justifying the generality of conclusions based on this loss. Second, to capture the characteristics of deep neural networks (DNNs), we introduce the notions of $\mathcal{H}(ψ)$-convexity and $\mathcal{H}(Ψ)$-smoothness, which generalize the classical concepts of strong convexity and Lipschitz smoothness. Based on them, we provide a theoretical explanation for the effectiveness of SGD in training DNNs. Finally, we derive model-independent bounds on the expected risk and generalization error for trained models, revealing the influence of the training set size, regularization term, the mutual information between labels and features, and the information loss caused by model irreversibility on risk and generalization. Based on our theoretical analysis and experimental validation, we believe that the PD learning framework facilitates a deeper and more unified theoretical understanding of deep learning.

Fan Xu, Nan Wang, Hao Wu et al.

Graph-Level Anomaly Detection (GLAD) aims to distinguish anomalous graphs within a graph dataset. However, current methods are constrained by their receptive fields, struggling to learn global features within the graphs. Moreover, most contemporary methods are based on spatial domain and lack exploration of spectral characteristics. In this paper, we propose a multi-perspective hybrid graph-level anomaly detector namely GLADformer, consisting of two key modules. Specifically, we first design a Graph Transformer module with global spectrum enhancement, which ensures balanced and resilient parameter distributions by fusing global features and spectral distribution characteristics. Furthermore, to uncover local anomalous attributes, we customize a band-pass spectral GNN message passing module that further enhances the model's generalization capability. Through comprehensive experiments on ten real-world datasets from multiple domains, we validate the effectiveness and robustness of GLADformer. This demonstrates that GLADformer outperforms current state-of-the-art models in graph-level anomaly detection, particularly in effectively capturing global anomaly representations and spectral characteristics.

Yuan Tai, Yihua Tan, Wei Gong et al.

The seven basic facial expression classifications are a basic way to express complex human emotions and are an important part of artificial intelligence research. Based on the traditional Bayesian neural network framework, the ResNet18_BNN network constructed in this paper has been improved in the following three aspects: (1) A new objective function is proposed, which is composed of the KL loss of uncertain parameters and the intersection of specific parameters. Entropy loss composition. (2) Aiming at a special objective function, a training scheme for alternately updating these two parameters is proposed. (3) Only model the parameters of the last convolution group. Through testing on the FER2013 test set, we achieved 71.5% and 73.1% accuracy in PublicTestSet and PrivateTestSet, respectively. Compared with traditional Bayesian neural networks, our method brings the highest classification accuracy gain.

Wei Gong, Laila Khalid

Machine learning is completely changing the trends in the fashion industry. From big to small every brand is using machine learning techniques in order to improve their revenue, increase customers and stay ahead of the trend. People are into fashion and they want to know what looks best and how they can improve their style and elevate their personality. Using Deep learning technology and infusing it with Computer Vision techniques one can do so by utilizing Brain-inspired Deep Networks, and engaging into Neuroaesthetics, working with GANs and Training them, playing around with Unstructured Data,and infusing the transformer architecture are just some highlights which can be touched with the Fashion domain. Its all about designing a system that can tell us information regarding the fashion aspect that can come in handy with the ever growing demand. Personalization is a big factor that impacts the spending choices of customers.The survey also shows remarkable approaches that encroach the subject of achieving that by divulging deep into how visual data can be interpreted and leveraged into different models and approaches. Aesthetics play a vital role in clothing recommendation as users' decision depends largely on whether the clothing is in line with their aesthetics, however the conventional image features cannot portray this directly. For that the survey also highlights remarkable models like tensor factorization model, conditional random field model among others to cater the need to acknowledge aesthetics as an important factor in Apparel recommendation.These AI inspired deep models can pinpoint exactly which certain style resonates best with their customers and they can have an understanding of how the new designs will set in with the community. With AI and machine learning your businesses can stay ahead of the fashion trends.

Wei Gong, Ningning Yan

In this paper we consider the convergence analysis of adaptive finite element method for elliptic optimal control problems with pointwise control constraints. We use variational discretization concept to discretize the control variable and piecewise linear and continuous finite elements to approximate the state variable. Based on the well-established convergence theory of AFEM for elliptic boundary value problems, we rigorously prove the convergence and quasi-optimality of AFEM for optimal control problems with respect to the state and adjoint state variables, by using the so-called perturbation argument. Numerical experiments confirm our theoretical analysis.

Taeho Jung, XuFei Mao, Xiang-Yang Li et al.

Much research has been conducted to securely outsource multiple parties' data aggregation to an untrusted aggregator without disclosing each individual's data, or to enable multiple parties to jointly aggregate their data while preserving privacy. However, those works either assume to have a secure channel or suffer from high complexity. Here we consider how an external aggregator or multiple parties learn some algebraic statistics (e.g., summation, product) over participants' data while any individual's input data is kept secret to others (the aggregator and other participants). We assume channels in our construction are insecure. That is, all channels are subject to eavesdropping attacks, and all the communications throughout the aggregation are open to others. We successfully guarantee data confidentiality under this weak assumption while limiting both the communication and computation complexity to at most linear.