Hui He

ICML Conference 2025 Conference Paper

Circumventing Backdoor Space via Weight Symmetry

• Jie Peng 0009
• Hongwei Yang
• Jing Zhao
• Hengji Dong
• Hui He
• Weizhe Zhang
• Haoyu He 0001

Deep neural networks are vulnerable to backdoor attacks, where malicious behaviors are implanted during training. While existing defenses can effectively purify compromised models, they typically require labeled data or specific training procedures, making them difficult to apply beyond supervised learning settings. Notably, recent studies have shown successful backdoor attacks across various learning paradigms, highlighting a critical security concern. To address this gap, we propose Two-stage Symmetry Connectivity (TSC), a novel backdoor purification defense that operates independently of data format and requires only a small fraction of clean samples. Through theoretical analysis, we prove that by leveraging permutation invariance in neural networks and quadratic mode connectivity, TSC amplifies the loss on poisoned samples while maintaining bounded clean accuracy. Experiments demonstrate that TSC achieves robust performance comparable to state-of-the-art methods in supervised learning scenarios. Furthermore, TSC generalizes to self-supervised learning frameworks, such as SimCLR and CLIP, maintaining its strong defense capabilities. Our code is available at https: //github. com/JiePeng104/TSC.

YNICL Journal 2025 Journal Article

Corrigendum to “Disturbed hierarchy and mediation in reward-related circuits in depression”. [45 (2025) 103739]

• Ruikun Yang
• Junxia Chen
• Suping Yue
• Yue Yu
• Jiamin Fan
• Yuling Luo
• Hui He
• Mingjun Duan

YNICL Journal 2025 Journal Article

Disturbed hierarchy and mediation in reward-related circuits in depression

• Ruikun Yang
• Junxia Chen
• Suping Yue
• Yue Yu
• Jiamin Fan
• Yuling Luo
• Hui He
• Mingjun Duan

BACKGROUNDS/OBJECTIVE: Deep brain stimulation (DBS) has proved the viability of alleviating depression symptoms by stimulating deep reward-related nuclei. This study aims to investigate the abnormal connectivity profiles among superficial, intermediate, and deep brain regions within the reward circuit in major depressive disorder (MDD) and therefore provides references for identifying potential superficial cortical targets for non-invasive neuromodulation. METHODS: Resting-state functional magnetic resonance imaging data were collected from a cohort of depression patients (N = 52) and demographically matched healthy controls (N = 60). Utilizing existing DBS targets as seeds, we conducted step-wise functional connectivity (sFC) analyses to delineate hierarchical pathways linking to cerebral cortices. Subsequently, the mediation effects of cortical regions on the interaction within reward-related circuits were further explored by constructing mediation models. RESULTS: In both cohorts, sFC analysis revealed two reward-related pathways from the deepest DBS targets to intermediate regions including the thalamus, insula, and anterior cingulate cortex (ACC), then to the superficial cortical cortex including medial frontal cortex, posterior default mode network (pDMN), and right dorsolateral prefrontal cortex (DLPFC). Patients exhibited reduced sFC in bilateral thalamus and medial frontal cortex in short and long steps respectively compared to healthy controls. We also discovered the disappearance of the mediation effects of superficial cortical regions on the interaction between DBS targets and intermediate regions in reward-related pathways in patients with MDD. CONCLUSION: Our findings support abnormal hierarchical connectivity and mediation effects in reward-related brain regions at different depth levels in MDD, which might elucidate the underlying pathophysiological mechanisms and inspire novel targets for non-invasive interventions.

NeurIPS Conference 2025 Conference Paper

SEMPO: Lightweight Foundation Models for Time Series Forecasting

• Hui He
• Kun Yi
• Yuanchi Ma
• Qi Zhang
• Zhendong Niu
• Guansong Pang

The recent boom of large pre-trained models witnesses remarkable success in developing foundation models (FMs) for time series forecasting. Despite impressive performance across diverse downstream forecasting tasks, existing time series FMs possess massive network architectures and require substantial pre-training on large-scale datasets, which significantly hinders their deployment in resource-constrained environments. In response to this growing tension between versatility and affordability, we propose SEMPO, a novel lightweight foundation model that requires pretraining on relatively small-scale data, yet exhibits strong general time series forecasting. Concretely, SEMPO comprises two key modules: 1) energy-aware S p E ctral decomposition module, that substantially improves the utilization of pre-training data by modeling not only the high-energy frequency signals but also the low-energy yet informative frequency signals that are ignored in current methods; and 2) M ixture-of- P r O mpts enabled Transformer, that learns heterogeneous temporal patterns through small dataset-specific prompts and adaptively routes time series tokens to prompt-based experts for parameter-efficient model adaptation across different datasets and domains. Equipped with these modules, SEMPO significantly reduces both pre-training data scale and model size, while achieving strong generalization. Extensive experiments on two large-scale benchmarks covering 16 datasets demonstrate the superior performance of SEMPO in both zero-shot and few-shot forecasting scenarios compared with state-of-the-art methods. Code and data are available at https: //github. com/mala-lab/SEMPO.

TAAS Journal 2025 Journal Article

SEPPDL: A Secure and Efficient Privacy-Preserving Deep Learning Inference Framework for Autonomous Driving

• Wang Bobo
• Hongwei Yang
• Meng Hao
• Jiannan Zhang
• Hui He
• Weizhe Zhang

The autonomous driving system necessitates using privacy-preserving deep learning (PPDL) technologies as the safety assurance for its extensive application. However, existing PPDL solutions depend on intricate protocol designs for robust security. Although leveraging advanced dedicated hardware platforms can significantly improve inference efficiency, the PPDL frameworks that make the best use of hardware platform computility are scarce. Thus, balancing efficiency and security in PPDL remains an open question. This study presents SEPPDL, a secure tripartite inference framework for deep learning based on secret-sharing to balance privacy security and computational efficiency. We reduce the communication and calculation time by designing a deep learning quantization representation scheme, two new computational protocols, and a computation library that utilizes the integer computation units of the GPU. The experimental results show that compared with state-of-the-art PPDL frameworks, the SEPPDL framework reduces the communication and computation delay in the model inference to 1/2 and 1/3 of the existing optimal frameworks while maintaining the accuracy of the model inference. Meanwhile, the SEPPDL framework achieves a 10-fold performance improvement in a lightweight model. As the model scale increases, the performance of the SEPPDL-based model even achieves an 86-fold improvement compared to VGG16.

NeurIPS Conference 2024 Conference Paper

FilterNet: Harnessing Frequency Filters for Time Series Forecasting

• Kun Yi
• Jingru Fei
• Qi Zhang
• Hui He
• Shufeng Hao
• Defu Lian
• Wei Fan

Given the ubiquitous presence of time series data across various domains, precise forecasting of time series holds significant importance and finds widespread real-world applications such as energy, weather, healthcare, etc. While numerous forecasters have been proposed using different network architectures, the Transformer-based models have state-of-the-art performance in time series forecasting. However, forecasters based on Transformers are still suffering from vulnerability to high-frequency signals, efficiency in computation, and bottleneck in full-spectrum utilization, which essentially are the cornerstones for accurately predicting time series with thousands of points. In this paper, we explore a novel perspective of enlightening signal processing for deep time series forecasting. Inspired by the filtering process, we introduce one simple yet effective network, namely FilterNet, built upon our proposed learnable frequency filters to extract key informative temporal patterns by selectively passing or attenuating certain components of time series signals. Concretely, we propose two kinds of learnable filters in the FilterNet: (i) Plain shaping filter, that adopts a universal frequency kernel for signal filtering and temporal modeling; (ii) Contextual shaping filter, that utilizes filtered frequencies examined in terms of its compatibility with input signals fordependency learning. Equipped with the two filters, FilterNet can approximately surrogate the linear and attention mappings widely adopted in time series literature, while enjoying superb abilities in handling high-frequency noises and utilizing the whole frequency spectrum that is beneficial for forecasting. Finally, we conduct extensive experiments on eight time series forecasting benchmarks, and experimental results have demonstrated our superior performance in terms of both effectiveness and efficiency compared with state-of-the-art methods. Our code is available at$^1$.

NeurIPS Conference 2023 Conference Paper

FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective

• Kun Yi
• Qi Zhang
• Wei Fan
• Hui He
• Liang Hu
• Pengyang Wang
• Ning An
• Longbing Cao

Multivariate time series (MTS) forecasting has shown great importance in numerous industries. Current state-of-the-art graph neural network (GNN)-based forecasting methods usually require both graph networks (e. g. , GCN) and temporal networks (e. g. , LSTM) to capture inter-series (spatial) dynamics and intra-series (temporal) dependencies, respectively. However, the uncertain compatibility of the two networks puts an extra burden on handcrafted model designs. Moreover, the separate spatial and temporal modeling naturally violates the unified spatiotemporal inter-dependencies in real world, which largely hinders the forecasting performance. To overcome these problems, we explore an interesting direction of directly applying graph networks and rethink MTS forecasting from a pure graph perspective. We first define a novel data structure, hypervariate graph, which regards each series value (regardless of variates or timestamps) as a graph node, and represents sliding windows as space-time fully-connected graphs. This perspective considers spatiotemporal dynamics unitedly and reformulates classic MTS forecasting into the predictions on hypervariate graphs. Then, we propose a novel architecture Fourier Graph Neural Network (FourierGNN) by stacking our proposed Fourier Graph Operator (FGO) to perform matrix multiplications in Fourier space. FourierGNN accommodates adequate expressiveness and achieves much lower complexity, which can effectively and efficiently accomplish {the forecasting}. Besides, our theoretical analysis reveals FGO's equivalence to graph convolutions in the time domain, which further verifies the validity of FourierGNN. Extensive experiments on seven datasets have demonstrated our superior performance with higher efficiency and fewer parameters compared with state-of-the-art methods. Code is available at this repository: https: //github. com/aikunyi/FourierGNN.

NeurIPS Conference 2023 Conference Paper

Frequency-domain MLPs are More Effective Learners in Time Series Forecasting

• Kun Yi
• Qi Zhang
• Wei Fan
• Shoujin Wang
• Pengyang Wang
• Hui He
• Ning An
• Defu Lian

Time series forecasting has played the key role in different industrial, including finance, traffic, energy, and healthcare domains. While existing literatures have designed many sophisticated architectures based on RNNs, GNNs, or Transformers, another kind of approaches based on multi-layer perceptrons (MLPs) are proposed with simple structure, low complexity, and superior performance. However, most MLP-based forecasting methods suffer from the point-wise mappings and information bottleneck, which largely hinders the forecasting performance. To overcome this problem, we explore a novel direction of applying MLPs in the frequency domain for time series forecasting. We investigate the learned patterns of frequency-domain MLPs and discover their two inherent characteristic benefiting forecasting, (i) global view: frequency spectrum makes MLPs own a complete view for signals and learn global dependencies more easily, and (ii) energy compaction: frequency-domain MLPs concentrate on smaller key part of frequency components with compact signal energy. Then, we propose FreTS, a simple yet effective architecture built upon Frequency-domain MLPs for Time Series forecasting. FreTS mainly involves two stages, (i) Domain Conversion, that transforms time-domain signals into complex numbers of frequency domain; (ii) Frequency Learning, that performs our redesigned MLPs for the learning of real and imaginary part of frequency components. The above stages operated on both inter-series and intra-series scales further contribute to channel-wise and time-wise dependency learning. Extensive experiments on 13 real-world benchmarks (including 7 benchmarks for short-term forecasting and 6 benchmarks for long-term forecasting) demonstrate our consistent superiority over state-of-the-art methods. Code is available at this repository: https: //github. com/aikunyi/FreTS.

AAAI Conference 2023 Conference Paper

InParformer: Evolutionary Decomposition Transformers with Interactive Parallel Attention for Long-Term Time Series Forecasting

• Haizhou Cao
• Zhenhao Huang
• Tiechui Yao
• Jue Wang
• Hui He
• Yangang Wang

Long-term time series forecasting (LTSF) provides substantial benefits for numerous real-world applications, whereas places essential demands on the model capacity to capture long-range dependencies. Recent Transformer-based models have significantly improved LTSF performance. It is worth noting that Transformer with the self-attention mechanism was originally proposed to model language sequences whose tokens (i.e., words) are discrete and highly semantic. However, unlike language sequences, most time series are sequential and continuous numeric points. Time steps with temporal redundancy are weakly semantic, and only leveraging time-domain tokens is hard to depict the overall properties of time series (e.g., the overall trend and periodic variations). To address these problems, we propose a novel Transformer-based forecasting model named InParformer with an Interactive Parallel Attention (InPar Attention) mechanism. The InPar Attention is proposed to learn long-range dependencies comprehensively in both frequency and time domains. To improve its learning capacity and efficiency, we further design several mechanisms, including query selection, key-value pair compression, and recombination. Moreover, InParformer is constructed with evolutionary seasonal-trend decomposition modules to enhance intricate temporal pattern extraction. Extensive experiments on six real-world benchmarks show that InParformer outperforms the state-of-the-art forecasting Transformers.

TIST Journal 2023 Journal Article

What Your Next Check-in Might Look Like: Next Check-in Behavior Prediction

• Heli Sun
• Chen Cao
• Xuguang Chu
• Tingting Hu
• Junzhi Lu
• Liang He
• Zhi Wang
• Hui He

In recent years, the next-POI recommendation has become a trending research topic in the field of trajectory data mining. For protection of user privacy, users’ complete GPS trajectories are difficult to obtain. The check-in information posted by users on social networks has become an important data source for Spatio-temporal Trajectory research. However, state-of-the-art methods neglect the social meaning and the information dissemination function of check-in behavior. The social meaning is an important reason why users are willing to post check-in on social networks, and the information dissemination function means, users can affect each other’s behavior by check-ins. The above characteristics of the check-in behavior make it different from the visiting behavior. We consider a new problem of predicting the next check-in behavior including the check-in time, the POI (point-of-interest) where the check-in is located, functional semantics of the POI, and so on. To solve the proposed problem, we build a multi-task learning model called DPMTM, and a pre-training module is designed to extract dynamic social semantics of check-in behaviors. Our results show that the DPMTM model works well in the check-in behavior problem.

AAAI Conference 2022 Conference Paper

CATN: Cross Attentive Tree-Aware Network for Multivariate Time Series Forecasting

• Hui He
• Qi Zhang
• Simeng Bai
• Kun Yi
• Zhendong Niu

Modeling complex hierarchical and grouped feature interaction in the multivariate time series data is indispensable to comprehending the data dynamics and predicting the future condition. The implicit feature interaction and highdimensional data make multivariate forecasting very challenging. Many existing works did not put more emphasis on exploring explicit correlation among multiple time-series data, and complicated models are designed to capture longand short-range patterns with the aid of attention mechanisms. In this work, we think that a pre-defined graph or a general learning method is difficult due to its irregular structure. Hence, we present CATN, an end-to-end model of Cross Attentive Tree-aware Network to jointly capture the interseries correlation and intra-series temporal patterns. We first construct a tree structure to learn hierarchical and grouped correlation and design an embedding approach that can pass a dynamic message to generalize implicit but interpretable cross features among multiple time series. Next in the temporal aspect, we propose a multi-level dependency learning mechanism including global&local learning and cross attention mechanism, which can combine long-range dependencies, short-range dependencies as well as cross dependencies at different time steps. The extensive experiments on different datasets from real-world show the effectiveness and robustness of the method we proposed when compared with existing state-of-the-art methods.

YNICL Journal 2019 Journal Article

BOLD-fMRI activity informed by network variation of scalp EEG in juvenile myoclonic epilepsy

• Yun Qin
• Sisi Jiang
• Qiqi Zhang
• Li Dong
• Xiaoyan Jia
• Hui He
• Yutong Yao
• Huanghao Yang

Epilepsy is marked by hypersynchronous bursts of neuronal activity, and seizures can propagate variably to any and all areas, leading to brain network dynamic organization. However, the relationship between the network characteristics of scalp EEG and blood oxygenation level-dependent (BOLD) responses in epilepsy patients is still not well known. In this study, simultaneous EEG and fMRI data were acquired in 18 juvenile myoclonic epilepsy (JME) patients. Then, the adapted directed transfer function (ADTF) values between EEG electrodes were calculated to define the time-varying network. The variation of network information flow within sliding windows was used as a temporal regressor in fMRI analysis to predict the BOLD response. To investigate the EEG-dependent functional coupling among the responding regions, modulatory interactions were analyzed for network variation of scalp EEG and BOLD time courses. The results showed that BOLD activations associated with high network variation were mainly located in the thalamus, cerebellum, precuneus, inferior temporal lobe and sensorimotor-related areas, including the middle cingulate cortex (MCC), supplemental motor area (SMA), and paracentral lobule. BOLD deactivations associated with medium network variation were found in the frontal, parietal, and occipital areas. In addition, modulatory interaction analysis demonstrated predominantly directional negative modulation effects among the thalamus, cerebellum, frontal and sensorimotor-related areas. This study described a novel method to link BOLD response with simultaneous functional network organization of scalp EEG. These findings suggested the validity of predicting epileptic activity using functional connectivity variation between electrodes. The functional coupling among the thalamus, frontal regions, cerebellum and sensorimotor-related regions may be characteristically involved in epilepsy generation and propagation, which provides new insight into the pathophysiological mechanisms and intervene targets for JME.

ECAI Conference 2016 Conference Paper

Randomized Canonical Correlation Discriminant Analysis for Face Recognition

• Bo Ma 0001
• Hui He
• Hongwei Hu
• Meili Wei

As an important technique in multivariate statistical analysis, Canonical Correlation Analysis (CCA) has been widely used in face recognition. But existing CCA based face recognition methods need two kinds of expression for the same face sample, and usually suffers high computational complexity in dealing with large samples. In this paper, we present a supervised method called Randomized Canonical Correlation Discriminant Analysis (RCCDA) based on Randomized non-linear Canonical Correlation Analysis (RCCA) to make up for the shortage of CCA based face recognition methods. We first obtain basis vectors approximately with random features instead of the calculation of kernel matrix to improve the efficiency of computation, then we use these basis vectors to compute random optimal discriminant features which can reduce the dimension of face features while preserving as much discriminatory information as possible. The result of experiments on Extended Yale B, AR, ORL and FERET face databases demonstrates that the performance of our method compares favorably with some state-of-the-art algorithms.

YNIMG Journal 2016 Journal Article

Structural and functional correlates of motor imagery BCI performance: Insights from the patterns of fronto-parietal attention network

• Tao Zhang
• Tiejun Liu
• Fali Li
• Mengchen Li
• Dongbo Liu
• Rui Zhang
• Hui He
• Peiyang Li

Motor imagery (MI)-based brain-computer interfaces (BCIs) have been widely used for rehabilitation of motor abilities and prosthesis control for patients with motor impairments. However, MI-BCI performance exhibits a wide variability across subjects, and the underlying neural mechanism remains unclear. Several studies have demonstrated that both the fronto-parietal attention network (FPAN) and MI are involved in high-level cognitive processes that are crucial for the control of BCIs. Therefore, we hypothesized that the FPAN may play an important role in MI-BCI performance. In our study, we recorded multi-modal datasets consisting of MI electroencephalography (EEG) signals, T1-weighted structural and resting-state functional MRI data for each subject. MI-BCI performance was evaluated using the common spatial pattern to extract the MI features from EEG signals. One cortical structural feature (cortical thickness (CT)) and two measurements (degree centrality (DC) and eigenvector centrality (EC)) of node centrality were derived from the structural and functional MRI data, respectively. Based on the information extracted from the EEG and MRI, a correlation analysis was used to elucidate the relationships between the FPAN and MI-BCI performance. Our results show that the DC of the right ventral intraparietal sulcus, the EC and CT of the left inferior parietal lobe, and the CT of the right dorsolateral prefrontal cortex were significantly associated with MI-BCI performance. Moreover, the receiver operating characteristic analysis and machine learning classification revealed that the EC and CT of the left IPL could effectively predict the low-aptitude BCI users from the high-aptitude BCI users with 83. 3% accuracy. Those findings consistently reveal that the individuals who have efficient FPAN would perform better on MI-BCI. Our findings may deepen the understanding of individual variability in MI-BCI performance, and also may provide a new biomarker to predict individual MI-BCI performance.