Author name cluster

Zhen Jia

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

2 author rows

ICML Conference 2025 Conference Paper

Proxsparse: Regularized Learning of Semi-Structured Sparsity masks for Pretrained LLMS

Hongyi Liu
Rajarshi Saha
Zhen Jia
Youngsuk Park
Jiaji Huang
Shoham Sabach
Yu-Xiang Wang 0003
George Karypis

Large Language Models (LLMs) have demonstrated exceptional performance in natural language processing tasks, yet their massive size makes serving them inefficient and costly. Semi-structured pruning has emerged as an effective method for model acceleration, but existing approaches are suboptimal because they focus on local, layer-wise optimizations using heuristic rules, failing to leverage global feedback. We present ProxSparse, a learning-based framework for mask selection enabled by regularized optimization. ProxSparse transforms the rigid, non-differentiable mask selection process into a smoother optimization procedure, allowing gradual mask exploration with flexibility. ProxSparse does not involve additional weight updates once the mask is determined. Our extensive evaluations on 7 widely used models show that ProxSparse consistently outperforms previously proposed semi-structured mask selection methods with significant improvement, demonstrating the effectiveness of our learned approach towards semi-structured pruning.

Details

EAAI Journal 2024 Journal Article

Intermittent fault diagnosis of analog circuit based on enhanced one-dimensional vision transformer and transfer learning strategy

Shengdong Wang
Zhenbao Liu
Zhen Jia
Wen Zhao
Zihao Li
Luyao Wang

As the major cause of false alarms in built-in test (BIT) system, intermittent faults of analog circuits may trigger abnormal equipment shutdown and lead to catastrophic accidents. With complete randomness and great non-repeatability, intermittent faults are arduous to be detected. To enhance the reliability and safety of electronic systems, an end-to-end approach based on enhanced one-dimensional Vision Transformer (1DViT) is proposed to realize intelligent diagnosis for intermittent faults of analog circuits. The signal anomaly caused by intermittent faults can be regarded as a kind of random anomaly from global perspective, and there are also rich local feature information in the fault interval. Completely composed of self-attention mechanism, Vision Transformer possesses prominent performance on extracting global features and modelling global representations, thus can be applied to identify intermittent faults. Meanwhile, to further enrich the feature representation, one multi-scale convolution fusion module (MSC) incorporating a series of convolution operations is designed and combined with 1DViT to extract and fuse the valuable local information. However, in practical test, due to the complex operation process, it is cumbersome to collect sufficient fault data to guarantee the effective training of the proposed model. To cope with this problem, transfer learning strategy is introduced. The model will be first pre-trained with adequate simulation data which is easily accessible, and then fine-tuned with a relatively small amount of actual fault data to help match the practical feature distribution. Experiments on two typical circuits demonstrate that the proposed method could achieve excellent diagnostic result in practical test.

Details DOI

EAAI Journal 2023 Journal Article

Incipient fault diagnosis of analog circuit with ensemble HKELM based on fused multi-channel and multi-scale features

Shengdong Wang
Zhenbao Liu
Zhen Jia
Zihao Li

As an essential part in electronics-rich system, the failure of analog circuits will severely affect the system reliability and security. Incipient fault of analog circuit refers to the early stage of degradation fault where the fault characteristics are generally weak and almost indistinguishable. In order to enhance the reliability of electronic systems, it is necessary to diagnose incipient faults of analog circuits promptly and effectively. Existing approaches generally capture fault characteristics only from single signal, ignoring the valuable information inherent in different domains and scales. To address this problem, a novel diagnostic strategy based on multi-scale feature extraction and multi-channel feature fusion is designed to guarantee the completeness and richness of fault information. In this study, a deep extreme learning machine denoising auto-encoder (DELM-DAE) based method is proposed to conduct unsupervised multi-scale and multi-channel feature fusion to extract distinguishable features for incipient faults. The proposed method has higher learning efficiency and overcomes the common problem of low efficiency in deep learning model training. Meanwhile, in order to improve the ability to distinguish high-resolution features, an ensemble hybrid kernel extreme learning machine with novel roulette selection and weighted voting scheme is proposed to enhance the recognition performance and stability. In the verification experiment, the diagnosis accuracy on four typical circuits all reaches above 98%, which demonstrates that the proposed incipient fault diagnosis method for analog circuits has more conspicuous performance than other state-of-the-art methods.

Details DOI

EAAI Journal 2023 Journal Article

Multi-scale integrated deep self-attention network for predicting remaining useful life of aero-engine

Ke Zhao
Zhen Jia
Feng Jia
Haidong Shao

Remaining useful life (RUL) prediction is the core research task of aero-engine prognostics health management (PHM), which is crucial to promoting the safety, reliability and economy. Therefore, in this paper, a multi-scale integrated deep self-attention network (MSIDSN) is proposed to process aero-engine multisensory data containing degradation information at different scales and then accurately predict the corresponding RUL of the aero-engine. Firstly, multi-scale blocks with self-attention strategy are constructed to selectively extract multisensory features on different scales. Secondly, an enhanced recurrent neural network module is designed to comprehensively extract the degraded features on multiple temporal scales. Finally, the feature fusion layer fuses the features and outputs the predicted RUL. Furthermore, an efficient loss function is developed to correct for the delay prediction, and avoid accidents. The experimental comparison results with other models verify the superiority of MSIDSN.

Details DOI

AAAI Conference 2018 Conference Paper

Deep Semantic Structural Constraints for Zero-Shot Learning

Yan Li
Zhen Jia
Junge Zhang
Kaiqi Huang
Tieniu Tan

Zero-shot learning aims to classify unseen image categories by learning a visual-semantic embedding space. In most cases, the traditional methods adopt a separated two-step pipeline that extracts image features from pre-trained CNN models. Then the ﬁxed image features are utilized to learn the embedding space. It leads to the lack of speciﬁc structural semantic information of image features for zero-shot learning task. In this paper, we propose an end-to-end trainable Deep Semantic Structural Constraints model to address this issue. The proposed model contains the Image Feature Structure constraint and the Semantic Embedding Structure constraint, which aim to learn structure-preserving image features and endue the learned embedding space with stronger generalization ability respectively. With the assistance of semantic structural information, the model gains more auxiliary clues for zero-shot learning. The state-of-the-art performance certiﬁes the effectiveness of our proposed method.

PDF Details

ICRA Conference 2011 Conference Paper

A two-step approach to see-through bad weather for surveillance video quality enhancement

Zhen Jia
Hongcheng Wang
Rodrigo E. Caballero
Ziyou Xiong
Jianwei Zhao
Alan Finn

Adverse weather conditions such as snow, fog or heavy rain greatly reduce the visual quality of outdoor surveillance videos. Video quality enhancement can improve the visual quality of surveillance videos providing clearer images with more details. Existing work in this area mainly focuses on quality enhancement for high resolution videos or still images, but few algorithms are developed for enhancing surveillance videos, which normally have low resolution, high noise and compression artifacts. In addition, for snow or rain conditions, the image quality of near-filed view is degraded by the obscuration of apparent snowflakes and raindrops, while the quality of far field view is degraded by the obscuration of fog-like snowflakes or raindrops. Very few video quality enhancement algorithms have been developed to handle both problems. In this paper, we propose a novel video quality enhancement algorithm for see-through snow, fog or heavy rain. The proposed algorithm has two major steps: 1. the near-field enhancement algorithm identifies obscuration pixels by snow or rain in the near-field view and removes these pixels as snowflakes or rain drops; different from state-of-the-art methods, the algorithm in this step can detect snowflakes on foreground object and background, and choose different methods to fill in the removed regions. 2. the far-field enhancement algorithm restores the image's contrast information not only to reveal more details in the far-field view but also to enhance the overall image's quality; in this step, the proposed algorithm adaptively enhances the global and local contrast, which is inspired on the human visual system, and accounts for the perceptual sensitivity to noise, compression artifacts, and the texture of image content. From our extensive testing, the proposed approach significantly improves the visual quality of surveillance videos by removing snow/fog/rain effects.

Details

ICRA Conference 2011 Conference Paper

Spatiotemporal energy modeling for foreground segmentation in multiple object tracking

Jie Shao 0010
Zhen Jia
Zhipeng Li
Fuqiang Liu
Jianwei Zhao
Pei-Yuan Peng

In this paper, we introduce spatiotemporal energy modeling for foreground segmentation in multiple object tracking, a high accuracy and real-time foreground target extraction algorithm. From a single video sequence with multiple moving objects and stationary background, our algorithm combines spatial (color distribution) and temporal (variety between two consecutive frames) information to extract foreground objects accurately and efficiently. The key idea of our method is to employ tracking results as feedback cues for target detection in the next frame, which adaptively updates the weights and threshold. Using spatiotemporal energy modeling, the foreground extraction errors caused by ambiguous colors in foreground and background boundary and abnormal movements can be substantially reduced. Experimental results of complex scenario video demonstrate the effectiveness of our algorithm.

Details

ICRA Conference 2005 Conference Paper

Sensor Fusion based 3D Target Visual Tracking for Autonomous Vehicles with IMM

Zhen Jia
Arjuna P. Balasuriya
Subhash Challa

This paper proposes an approach for object identification and tracking for autonomous vehicle application. In this scheme, data from the vehicle’s onboard vision and motion sensors are fused to identify the target 3D dynamic features in the world coordinate. Here several simple and basic linear dynamic models are combined to make the approximation of the target’s unpredicted or complex motion properties. With these basic linear dynamic models a detailed description of the 3D target tracking system with the interacting multiple models (IMM) for Extended Kalman Filtering is presented. The target’s final state estimates are obtained as a weighted combination of the outputs from each different model. Performance of the proposed interacting multiple dynamic model tracking algorithm is demonstrated through experimental results.

Details