Yaohua Liu (刘耀华)

Yaohua Liu (刘耀华)

Ph.D.

Publications

You can also find my articles on my DBLP profile.

Conference Papers

Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncation

Published in International Joint Conference on Artificial Intelligence (IJCAI), 2024

Transfer attacks generate significant interest for real-world black-box applications by crafting transferable adversarial examples through surrogate models. Whereas, existing works essentially directly optimize the single-level objective w.r.t. the surrogate model, which always leads to poor interpretability of attack mechanism and limited generalization performance over unknown victim models. …

Recommended citation: Yaohua Liu, Jiaxin Gao, Xuan Liu, Xianghao Jiao, Xin Fan, Risheng Liu. Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncation[C]. International Joint Conference on Artificial Intelligence (IJCAI), 2024.
Download Paper | Code Repository

Enhancing images with coupled low-resolution and ultra-dark degradations: a tri-level learning framework

Published in ACM International Conference on Multimedia (ACM MM), 2024

Due to device constraints and lighting conditions, captured images frequently exhibit coupled low-resolution and ultra-dark degradations. Enhancing the visibility and resolution of ultra-dark images simultaneously is crucial for practical applications. …

Recommended citation: Jiaxin Gao, Yaohua Liu*. Enhancing images with coupled low-resolution and ultra-dark degradations: a tri-level learning framework[C]. ACM International Conference on Multimedia (ACM MM), 2024.
Download Paper | Code Repository

Motion-Scenario Decoupling for Rat-Aware Video Position Prediction: Strategy and Benchmark

Published in International Conference on Image and Graphics (ICIG), 2023

Recently significant progress has been made in human action recognition and behavior prediction using deep learning techniques, leading to improved vision-based semantic understanding. However, there is still a lack of high-quality motion datasets for small bio-robotics, which presents more challenging scenarios for long-term movement prediction and behavior control based on third-person observation. …

Recommended citation: Xiaofeng Liu, Jiaxin Gao, Yaohua Liu, et al. Motion-Scenario Decoupling for Rat-Aware Video Position Prediction: Strategy and Benchmark[C]. International Conference on Image and Graphics (ICIG), 2023, Oral.
Download Paper

PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentation

Published in ACM International Conference on Multimedia (ACM MM), 2023

In light of the significant progress made in the development and application of semantic segmentation tasks, there has been increasing attention towards improving the robustness of segmentation models against natural degradation factors (e.g., rain streaks) or artificially attack factors (e.g., adversarial attack). …

Recommended citation: Xianghao Jiao, Yaohua Liu, Jiaxin Gao, et al. Pearl: Preprocessing enhanced adversarial robust learning of image deraining for semantic segmentation[C]. ACM International Conference on Multimedia (ACM MM), 2023.
Download Paper

Averaged Method of Multipliers for Bi-Level Optimization without Lower-Level Strong Convexity

Published in International Conference on Machine Learning (ICML), 2023

Gradient methods have become mainstream techniques for Bi-Level Optimization (BLO) in learning fields. The validity of existing works heavily rely on either a restrictive Lower- Level Strong Convexity (LLSC) condition or on solving a series of approximation subproblems with high accuracy or both. …

Recommended citation: Risheng Liu, Yaohua Liu, Wei Yao, Shangzhi Zeng, Jin Zhang. Averaged Method of Multipliers for Bi-Level Optimization without Lower-Level Strong Convexity[C]. International Conference on Machine Learning (ICML), 2023.
Code Repository

Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyonds

Published in Advances in Neural Information Processing Systems 34 (NeurIPS), 2021

In recent years, Bi-Level Optimization (BLO) techniques have received extensive attentions from both learning and vision communities. A variety of BLO models in complex and practical tasks are of non-convex follower structure in nature (a.k.a., without Lower-Level Convexity, LLC for short). …

Recommended citation: Risheng Liu, Yaohua Liu, Shangzhi Zeng, Jin Zhang. Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyond[C]. Advances in Neural Information Processing Systems (NeurIPS), 2021, Spotlight, Acceptance Rate ≤ 3% .
Download Paper | Code Repository

Boml: A Modularized Bilevel Optimization Library In Python For Meta Learning

Published in IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 2021

Meta-learning (a.k.a. learning to learn) has recently emerged as a promising paradigm for a variety of applications. There are now many meta-learning methods, each focusing on different modeling aspects of base and meta learners, but all can be (re)formulated as specific bilevel optimization problems. …

Recommended citation: Yaohua Liu, Risheng Liu. Boml: A Modularized Bilevel Optimization Library In Python For Meta Learning[C]. IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 2021, Best Open Source Project Awards, Oral.
Download Paper | Code Repository

Journal Articles

Augmenting Iterative Trajectory for Bilevel Optimization: Methodology, Analysis and Extensions

Published in Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2025

Abstract—In recent years, there has been a surge of machine learning applications developed with hierarchical structure, which can be approached from Bi-Level Optimization (BLO) perspective. However, most existing gradient-based methods overlook the interdependence between hyper-gradient calculation and Lower-Level (LL) iterative trajectory, focusing solely on the former. Consequently, convergence theory is constructed with restrictive LL assumptions, which are often challenging to satisfy in real-world scenarios. In this work, …

Recommended citation: Risheng Liu, Yaohua Liu, Shangzhi Zeng, Jin Zhang. Augmenting Iterative Trajectory for Bilevel Optimization: Methodology, Analysis and Extensions[J]. Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2025.
Download Paper | Code Repository

Collaborative brightening and amplification of low-light imagery via bi-level adversarial learning

Published in Pattern Recognition (PR), 2024

Poor light conditions constrain the high pursuit of clarity and visible quality of photography especially smartphone devices. Admittedly, existing specific image processing methods, whether super-resolution methods or low-light enhancement methods, can hardly simultaneously enhance the resolution and brightness of low-light images at the same time. …

Recommended citation: Jiaxin Gao, Yaohua Liu, Ziyu Yue, et al. Collaborative brightening and amplification of low-light imagery via bi-level adversarial learning[J]. Pattern Recognition (PR), 2024, 154: 110558.
Download Paper | Code Repository

A Dual-Stream-Modulated Learning Framework for Illuminating and Super-Resolving Ultra-Dark Images

Published in IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS), 2024

Enhancement of image resolution for scenes captured under extremely dim conditions represents a practical yet challenging problem that has received little attention. In such low-light scenarios, the limited lighting and minimal signal clarity tend to intensify issues such as diminished detail visibility and altered color accuracy, which are often more severe during the image enhancement process than in scenarios with adequate lighting. …

Recommended citation: Jiaxin Gao, Ziyu Yue, Yaohua Liu, et al. A Dual-Stream-Modulated Learning Framework for Illuminating and Super-Resolving Ultra-Dark Images[J]. IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS), 2024.
Download Paper | Code Repository

Triple-level model inferred collaborative network architecture for video deraining

Published in IEEE Transactions on Image Processing (IEEE TIP), 2021

Video deraining is an important issue for outdoor vision systems and has been investigated extensively. However, designing optimal architectures by the aggregating model formation and data distribution is a challenging task for video deraining. …

Recommended citation: Pan Mu, Zhu Liu, Yaohua Liu, Risheng Liu, Xin Fan. Triple-level model inferred collaborative network architecture for video deraining[J]. IEEE Transactions on Image Processing (IEEE TIP), 2021, 154: 110558.
Download Paper