Author name cluster

Changyuan Yang

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

2 author rows

ICLR Conference 2025 Conference Paper

Distribution Backtracking Builds A Faster Convergence Trajectory for Diffusion Distillation

Shengyuan Zhang
Ling Yang 0006
Zejian Li
An Zhao
Chenye Meng
Changyuan Yang
Guang Yang 0022
Zhiyuan Yang

Accelerating the sampling speed of diffusion models remains a significant challenge. Recent score distillation methods distill a heavy teacher model into a student generator to achieve one-step generation, which is optimized by calculating the difference between two score functions on the samples generated by the student model. However, there is a score mismatch issue in the early stage of the score distillation process, since existing methods mainly focus on using the endpoint of pre-trained diffusion models as teacher models, overlooking the importance of the convergence trajectory between the student generator and the teacher model. To address this issue, we extend the score distillation process by introducing the entire convergence trajectory of the teacher model and propose $\textbf{Dis}$tribution $\textbf{Back}$tracking Distillation ($\textbf{DisBack}$). DisBask is composed of two stages: $\textit{Degradation Recording}$ and $\textit{Distribution Backtracking}$. $\textit{Degradation Recording}$ is designed to obtain the convergence trajectory by recording the degradation path from the pre-trained teacher model to the untrained student generator. The degradation path implicitly represents the intermediate distributions between the teacher and the student, and its reverse can be viewed as the convergence trajectory from the student generator to the teacher model. Then $\textit{Distribution Backtracking}$ trains the student generator to backtrack the intermediate distributions along the path to approximate the convergence trajectory of the teacher model. Extensive experiments show that DisBack achieves faster and better convergence than the existing distillation method and achieves comparable or better generation performance, with an FID score of 1.38 on the ImageNet 64$\times$64 dataset. DisBack is easy to implement and can be generalized to existing distillation methods to boost performance.

Details

AAAI Conference 2024 Conference Paper

Reducing Spatial Fitting Error in Distillation of Denoising Diffusion Models

Shengzhe Zhou
Zejian Li
Shengyuan Zhang
Lefan Hou
Changyuan Yang
Guang Yang
Zhiyuan Yang
Lingyun Sun

Denoising Diffusion models have exhibited remarkable capabilities in image generation. However, generating high-quality samples requires a large number of iterations. Knowledge distillation for diffusion models is an effective method to address this limitation with a shortened sampling process but causes degraded generative quality. Based on our analysis with bias-variance decomposition and experimental observations, we attribute the degradation to the spatial fitting error occurring in the training of both the teacher and student model in the distillation. Accordingly, we propose Spatial Fitting-Error Reduction Distillation model (SFERD). SFERD utilizes attention guidance from the teacher model and a designed semantic gradient predictor to reduce the student's fitting error. Empirically, our proposed model facilitates high-quality sample generation in a few function evaluations. We achieve an FID of 5.31 on CIFAR-10 and 9.39 on ImageNet 64x64 with only one step, outperforming existing diffusion methods. Our study provides a new perspective on diffusion distillation by highlighting the intrinsic denoising ability of models.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Learning Object Consistency and Interaction in Image Generation from Scene Graphs

Yangkang Zhang
Chenye Meng
Zejian Li
Pei Chen
Guang Yang
Changyuan Yang
Lingyun Sun

This paper is concerned with synthesizing images conditioned on a scene graph (SG), a set of object nodes and their edges of interactive relations. We divide existing works into image-oriented and code-oriented methods. In our analysis, the image-oriented methods do not consider object interaction in spatial hidden feature. On the other hand, in empirical study, the code-oriented methods lose object consistency as their generated images miss certain objects in the input scene graph. To alleviate these two issues, we propose Learning Object Consistency and Interaction (LOCI). To preserve object consistency, we design a consistency module with a weighted augmentation strategy for objects easy to be ignored and a matching loss between scene graphs and image codes. To learn object interaction, we design an interaction module consisting of three kinds of message propagation between the input scene graph and the learned image code. Experiments on COCO-stuff and Visual Genome datasets show our proposed method alleviates the ignorance of objects and outperforms the state-of-the-art on visual fidelity of generated images and objects.

PDF Details DOI

EAAI Journal 2023 Journal Article

Robust discriminant latent variable manifold learning for rotating machinery fault diagnosis

Changyuan Yang
Sai Ma
Qinkai Han

Fault diagnosis is an important technology for performing intelligent manufacturing. To simultaneously maintain high manufacturing quality and low failure rate for manufacturing systems, it is of great value to accurately locate the fault element, evaluate the fault severity and find the fault root cause. In order to effectively perform feature extraction for rotating machinery fault diagnosis, a robust discriminant latent variable manifold learning (RDLVML) algorithm is proposed in this paper. Specifically, RDLVML algorithm could effectively select the features of the high-dimensional fault data, and extract the low-dimensional fault features with better discrimination. Through the idea of discriminative learning, the accuracy of fault diagnosis was improved by increasing the aggregation of the same state feature samples and the difference between different classes of feature samples. A novel weighted neighborhood graph is proposed by constructing the q-Rényi kernel function, which could effectively suppress the interference of outliers and noise, and make the RDLVML algorithm more robust. Moreover, the fault diagnosis method of rotating machinery based on RDLVML is proposed in this paper. Through experimental verifications and comparisons with several classical feature selection algorithms, rotating machinery fault diagnosis could be more accurately performed by this method.

Details DOI