Author name cluster

Ming Wu

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.

8 papers

1 author row

EAAI Journal 2025 Journal Article

Data-driven models with physical interpretability for real-time cavity profile prediction in electrochemical machining processes

Ming Wu
Zequan Yao
Mathias Verbeke
Peter Karsmakers
Benjamin Gorissen
Dominiek Reynaerts

The Electrochemical Machining (ECM) process can be effectively controlled by adjusting the process parameters, including electrolyte composition, current density, etc. However, this control strategy presents challenges associated with modeling the complex relationships between processing parameters and the resulting process outcomes. Data-driven approaches hold promise for real-time cavity profile prediction, while their black-box nature limits the interpretability. To address these issues, this study initiated by developing several Machine Learning (ML) models to predict cavity profiles using process parameters, in-process data, or a combination of both. Subsequently, these models were inspected by explainable artificial intelligence (XAI) methods. Using process parameters for model inspection provides preliminary guidance for achieving desired outcomes, while inspections with in-process data shed light on process dynamics and enable diagnostics by detecting anomalies through atypical model focus. Linear regression (LR), with high interpretability, offers moderate predictive accuracy, whereas neural networks (NN) and convolutional neural networks (CNN) perform better yet require XAI methods to interpret their decision-making mechanisms. Model inspection was conducted on global and local levels. Global inspection via SHapley Additive exPlanations (SHAP) for LR and NN models identified factors affecting cavity size, aligning predictions with ECM knowledge. Local inspection through Gradient-weighted Class Activation Mapping (Grad-CAM) examined CNN predictions, revealing temporal process dynamics and model focus during specific processing stages. The impact of unexpected events identified by the ML models was validated using a physics-based model. A web application was developed to integrate these insights, allowing for real-time cavity prediction and visualization based on process parameters and in-process data.

Details DOI

EAAI Journal 2024 Journal Article

Label noise correction for crowdsourcing using dynamic resampling

Jing Zhang
Xiaoqian Jiang
Nianshang Tian
Ming Wu

Crowdsourcing provides a cost-effective labeling solution for the acquisition of labeled training samples for machine learning by employing workers on the Internet. A common approach to improving the label quality is to employ a truth inference method to infer integrated labels for samples from their multiple noisy labels obtained from different crowd workers. Although the quality of integrated labels is significantly improved compared with that of the original noisy ones, it still cannot completely eliminate the noises inevitably existing in the integrated labels. To further improve the label quality, this paper proposes a novel label noise correction method for crowdsourcing based on dynamic resampling (DRNC). DRNC first divides the dataset with inferred labels into a clean set and a noisy set through a filter. According to a certain proportion, the clean set and the noisy set are resampled to train multiple heterogeneous classifiers, which form an ensemble classifier. Then, the dataset is divided by the ensemble classifier into a new sub-noisy set and a sub-clean set. The whole process repeats multiple rounds, generating multiple sub-clean sets. Finally, these sub-clean sets are used to train classifiers, which jointly correct the wrong labels in the dataset by voting. Experimental results on 25 simulated and 4 real-world datasets consistently show that the proposed DRNC averagely improves the quality of labels as well as the quality of learned models in the range of 1 to 10 percentage points, compared with four state-of-the-art crowdsourcing noise correction methods.

Details DOI

AAAI Conference 2024 Conference Paper

Privileged Prior Information Distillation for Image Matting

Cheng Lyu
Jiake Xie
Bo Xu
Cheng Lu
Han Huang
Xin Huang
Ming Wu
Chuang Zhang

Performance of trimap-free image matting methods is limited when trying to decouple the deterministic and undetermined regions, especially in the scenes where foregrounds are semantically ambiguous, chromaless, or high transmittance. In this paper, we propose a novel framework named Privileged Prior Information Distillation for Image Matting (PPID-IM) that can effectively transfer privileged prior environment-aware information to improve the performance of trimap-free students in solving hard foregrounds. The prior information of trimap regulates only the teacher model during the training stage, while not being fed into the student network during actual inference. To achieve effective privileged cross-modality (i.e. trimap and RGB) information distillation, we introduce a Cross-Level Semantic Distillation (CLSD) module that reinforces the students with more knowledgeable semantic representations and environment-aware information. We also propose an Attention-Guided Local Distillation module that efficiently transfers privileged local attributes from the trimap-based teacher to trimap-free students for the guidance of local-region optimization. Extensive experiments demonstrate the effectiveness and superiority of our PPID on image matting. The code will be released soon.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Doubly-Robust Self-Training

Banghua Zhu
Mingyu Ding
Philip Jacobson
Ming Wu
Wei Zhan
Michael Jordan
Jiantao Jiao

Self-training is a well-established technique in semi-supervised learning, which leverages unlabeled data by generating pseudo-labels and incorporating them with a limited labeled dataset for training. The effectiveness of self-training heavily relies on the accuracy of these pseudo-labels. In this paper, we introduce doubly-robust self-training, an innovative semi-supervised algorithm that provably balances between two extremes. When pseudo-labels are entirely incorrect, our method reduces to a training process solely using labeled data. Conversely, when pseudo-labels are completely accurate, our method transforms into a training process utilizing all pseudo-labeled data and labeled data, thus increasing the effective sample size. Through empirical evaluations on both the ImageNet dataset for image classification and the nuScenes autonomous driving dataset for 3D object detection, we demonstrate the superiority of the doubly-robust loss over the self-training baseline.

PDF Details

AAAI Conference 2023 Conference Paper

SwiftAvatar: Efficient Auto-Creation of Parameterized Stylized Character on Arbitrary Avatar Engines

Shizun Wang
Weihong Zeng
Xu Wang
Hao Yang
Li Chen
Chuang Zhang
Ming Wu
Yi Yuan

The creation of a parameterized stylized character involves careful selection of numerous parameters, also known as the "avatar vectors" that can be interpreted by the avatar engine. Existing unsupervised avatar vector estimation methods that auto-create avatars for users, however, often fail to work because of the domain gap between realistic faces and stylized avatar images. To this end, we propose SwiftAvatar, a novel avatar auto-creation framework that is evidently superior to previous works. SwiftAvatar introduces dual-domain generators to create pairs of realistic faces and avatar images using shared latent codes. The latent codes can then be bridged with the avatar vectors as pairs, by performing GAN inversion on the avatar images rendered from the engine using avatar vectors. Through this way, we are able to synthesize paired data in high-quality as many as possible, consisting of avatar vectors and their corresponding realistic faces. We also propose semantic augmentation to improve the diversity of synthesis. Finally, a light-weight avatar vector estimator is trained on the synthetic pairs to implement efficient auto-creation. Our experiments demonstrate the effectiveness and efficiency of SwiftAvatar on two different avatar engines. The superiority and advantageous flexibility of SwiftAvatar are also verified in both subjective and objective evaluations.

PDF Details DOI

EAAI Journal 2022 Journal Article

Self-supervised monocular depth estimation in dynamic scenes with moving instance loss

Min Yue
Guangyuan Fu
Ming Wu
Xin Zhang
Hongyang Gu

Estimating depth from monocular images is a powerful method to perceive valuable environmental information, which is essential for applications that require three-dimensional (3D) environmental models such as autonomous driving and virtual reality. The monocular self-supervised depth estimation method based on deep learning has made rapid progress without depth ground truth information. However, the existing methods are based on the assumption of a static world during training, and depth estimation in a dynamic environment needs further development. To solve this problem, we propose a new monocular self-supervised depth estimation method in dynamic scenes to eliminate the negative impact of moving objects in the image sequence when calculating the self-supervised loss. Specifically, for the self-supervised depth estimation framework, we propose a moving object mask based on the minimum instance photometric residual and then combine it with the mask based on instance re-projection residual in the existing instance-level moving object segmentation methods. In addition, we design a moving instance loss function to process the moving object, so that the training of the model can achieve better performance. Experiments are conducted on public datasets to verify the effectiveness of the proposed method and each of its components, and the results show that our method achieves better performance for depth estimation in dynamic scenes compared to state-of-the-art methods.

Details DOI

AAAI Conference 2020 Conference Paper

Learning Feature Interactions with Lorentzian Factorization Machine

Canran Xu
Ming Wu

Learning representations for feature interactions to model user behaviors is critical for recommendation system and click-trough rate (CTR) predictions. Recent advances in this area are empowered by deep learning methods which could learn sophisticated feature interactions and achieve the stateof-the-art result in an end-to-end manner. These approaches require large number of training parameters integrated with the low-level representations, and thus are memory and computational inefﬁcient. In this paper, we propose a new model named “LorentzFM” that can learn feature interactions embedded in a hyperbolic space in which the violation of triangle inequality for Lorentz distances is available. To this end, the learned representation is beneﬁted by the peculiar geometric properties of hyperbolic triangles, and result in a signiﬁcant reduction in the number of parameters (20% to 80%) because all the top deep learning layers are not required. With such a lightweight architecture, LorentzFM achieves comparable and even materially better results than the deep learning methods such as DeepFM, xDeepFM and Deep & Cross in both recommendation and CTR prediction tasks.

PDF Details

EAAI Journal 2018 Journal Article

Data-driven fault prediction and anomaly measurement for complex systems using support vector probability density estimation

Hong-Qiao Wang
Yan-Ning Cai
Guang-Yuan Fu
Ming Wu
Zhen-Hua Wei

To quantitatively monitor the state of complex system, a data-driven fault prediction and anomaly degree measurement method based on probability density estimation is studied in this paper. First, an anomaly index is introduced and defined to measure the anomaly degree of samples. Then By improving the form of constraint condition, a single slack factor multiple kernel support vector machine probability density estimation model is presented. As a result, the scale of object function and the solution number are all reduced, and the computational efficiency of the presented model is greatly enhanced. On the other hand, as the introduction of multiple kernel functions, a multiple kernel matrix with better data mapping performance is obtained, which can well solve the composite probability density estimation for uncoupled data. The simulation test shows that the presented model has higher estimation precision and speed. The experiments on complex system fault prediction also show that the system’s anomaly degree can be quantitatively and accurately measured by the anomaly index gained from the prediction results, which can effectively improve the fault prediction precision and increase the prediction advances.

Details DOI