Author name cluster

Lei Cai

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

EAAI Journal 2026 Journal Article

Machine-learning-based pattern recognition and key discharge mode diagnosis of multi-source discharge in a real switch cabinet: A perspective from optical signals

Hongtu Cheng
Yang Shen
Qi Hu
Jie Feng
Lei Cai
Xi Zhu
Zhi Fang

Partial discharge (PD) in gas-insulated switch cabinets exhibits spatial distribution, multiple sources, and diverse types, posing challenges for fault diagnosis. The existing detection and pattern recognition methods are unable to effectively address the issues of distributed detection and signal interference that arise when PD co-occurs at multiple locations within the equipment. Fluorescence optical fiber was employed to acquire optical signals corresponding to four typical types of PDs and their 15 spatially distributed multi-source combinations in a real switch cabinet. Discharge data for 15 types, with 1000 cycles per type, totaling 15000 cycles, were collected. The dataset was split into training and testing sets at an 8: 2 ratio, ensuring no overlap between them. To accurately classify the complex discharge modes, we developed a machine learning-based pattern recognition model. This model integrates extreme gradient boosting (XGB) and light gradient boosting machine (LGBM) via a soft voting ensemble, following Bayesian hyperparameter optimization. A dynamic weight adjustment mechanism was also incorporated to address class imbalance. Results demonstrate that the integrated model achieved an overall classification accuracy of 93%. The number of fault types with a diagnostic rate below 90% decreased from 4 to 2, with the diagnostic rate for all fault types exceeding 80%. Recall and F1 scores for most categories remain above 90%, indicating strong performance in identifying multiple types of PD under realistic operational conditions. This research provides a reference for real-time monitoring and intelligent early warning of multi-source PDs in power equipment.

Details DOI

AAAI Conference 2020 Conference Paper

A Multi-Scale Approach for Graph Link Prediction

Lei Cai
Shuiwang Ji

Deep models can be made scale-invariant when trained with multi-scale information. Images can be easily made multi-scale, given their grid-like structures. Extending this to generic graphs poses major challenges. For example, in link prediction tasks, inputs are represented as graphs consisting of nodes and edges. Currently, the state-of-the-art model for link prediction uses supervised heuristic learning, which learns graph structure features centered on two target nodes. It then learns graph neural networks to predict the existence of links based on graph structure features. Thus, the performance of link prediction models highly depends on graph structure features. In this work, we propose a novel node aggregation method that can transform the enclosing subgraph into different scales and preserve the relationship between two target nodes for link prediction. A theory for analyzing the information loss during the re-scaling procedure is also provided. Graphs in different scales can provide scaleinvariant information, which enables graph neural networks to learn invariant features and improve link prediction performance. Our experimental results on 14 datasets from different areas demonstrate that our proposed method outperforms the state-of-the-art methods by employing multi-scale graphs without additional parameters.

PDF Details

AAAI Conference 2020 Conference Paper

Adaptive Convolutional ReLUs

Hongyang Gao
Lei Cai
Shuiwang Ji

Rectiﬁed linear units (ReLUs) are currently the most popular activation function used in neural networks. Although ReLUs can solve the gradient vanishing problem and accelerate training convergence, it suffers from the dying ReLU problem in which some neurons are never activated if the weights are not updated properly. In this work, we propose a novel activation function, known as the adaptive convolutional ReLU (ConvReLU), that can better mimic brain neuron activation behaviors and overcome the dying ReLU problem. With our novel parameter sharing scheme, ConvReLUs can be applied to convolution layers that allow each input neuron to be activated by different trainable thresholds without involving a large number of extra parameters. We employ the zero initialization scheme in ConvReLU to encourage trainable thresholds to be close to zero. Finally, we develop a partial replacement strategy that only replaces the ReLUs in the early layers of the network. This resolves the dying ReLU problem and retains sparse representations for linear classiﬁers. Experimental results demonstrate that our proposed ConvReLU has consistently better performance compared to ReLU, LeakyReLU, and PReLU. In addition, the partial replacement strategy is shown to be effective not only for our ConvReLU but also for LeakyReLU and PReLU.

PDF Details

IJCAI Conference 2019 Conference Paper

Dense Transformer Networks for Brain Electron Microscopy Image Segmentation

Jun Li
Yongjun Chen
Lei Cai
Ian Davidson
Shuiwang Ji

The key idea of current deep learning methods for dense prediction is to apply a model on a regular patch centered on each pixel to make pixel-wise predictions. These methods are limited in the sense that the patches are determined by network architecture instead of learned from data. In this work, we propose the dense transformer networks, which can learn the shapes and sizes of patches from data. The dense transformer networks employ an encoder-decoder architecture, and a pair of dense transformer modules are inserted into each of the encoder and decoder paths. The novelty of this work is that we provide technical solutions for learning the shapes and sizes of patches from data and efficiently restoring the spatial correspondence required for dense prediction. The proposed dense transformer modules are differentiable, thus the entire network can be trained. We apply the proposed networks on biological image segmentation tasks and show superior performance is achieved in comparison to baseline methods.

PDF Details