Congying Han
YOU?
Author Swipe
View article: A Mutil-conditional Diffusion Transformer for Versatile Seismic Wave Generation
A Mutil-conditional Diffusion Transformer for Versatile Seismic Wave Generation Open
Seismic wave generation creates labeled waveform datasets for source parameter inversion, subsurface analysis, and, notably, training artificial intelligence seismology models. Traditionally, seismic wave generation has been time-consuming…
View article: Purity Law for Generalizable Neural TSP Solvers
Purity Law for Generalizable Neural TSP Solvers Open
Achieving generalization in neural approaches across different scales and distributions remains a significant challenge for the Traveling Salesman Problem~(TSP). A key obstacle is that neural networks often fail to learn robust principles …
View article: StyO: Stylize Your Face in Only One-Shot
StyO: Stylize Your Face in Only One-Shot Open
This paper focuses on face stylization with a single artistic target. Existing works for this task often fail to retain the source content while achieving geometry variation. Here, we present a novel StyO model, i.e., Stylize the face in o…
View article: Towards Optimal Adversarial Robust Reinforcement Learning with Infinity Measurement Error
Towards Optimal Adversarial Robust Reinforcement Learning with Infinity Measurement Error Open
Ensuring the robustness of deep reinforcement learning (DRL) agents against adversarial attacks is critical for their trustworthy deployment. Recent research highlights the challenges of achieving state-adversarial robustness and suggests …
View article: Dual Alignment Maximin Optimization for Offline Model-based RL
Dual Alignment Maximin Optimization for Offline Model-based RL Open
Offline reinforcement learning agents face significant deployment challenges due to the synthetic-to-real distribution mismatch. While most prior research has focused on improving the fidelity of synthetic sampling and incorporating off-po…
View article: Preference-based opponent shaping in differentiable games
Preference-based opponent shaping in differentiable games Open
Strategy learning in game environments with multi-agent is a challenging problem. Since each agent's reward is determined by the joint strategy, a greedy learning strategy that aims to maximize its own reward may fall into a local optimum.…
View article: DR-BFR: Degradation Representation with Diffusion Models for Blind Face Restoration
DR-BFR: Degradation Representation with Diffusion Models for Blind Face Restoration Open
Blind face restoration (BFR) is fundamentally challenged by the extensive range of degradation types and degrees that impact model generalization. Recent advancements in diffusion models have made considerable progress in this field. Never…
View article: Mitigating Distribution Shift in Model-based Offline RL via Shifts-aware Reward Learning
Mitigating Distribution Shift in Model-based Offline RL via Shifts-aware Reward Learning Open
Model-based offline reinforcement learning trains policies using pre-collected datasets and learned environment models, eliminating the need for direct real-world environment interaction. However, this paradigm is inherently challenged by …
View article: Cassava Starch-Based Multifunctional Coating Incorporated with Zinc Oxide Nanoparticle to Enhance the Shelf Life of Passion Fruit
Cassava Starch-Based Multifunctional Coating Incorporated with Zinc Oxide Nanoparticle to Enhance the Shelf Life of Passion Fruit Open
Passion fruits are susceptible to numerous postharvest challenges including weight loss, ethylene production, peel shrinkage, microbial growth, and pulp liquefaction. To mitigate these issues, yellow passion fruits were treated with hydrox…
View article: Identification of novel variations in three cases with rare inherited neuromuscular disorder
Identification of novel variations in three cases with rare inherited neuromuscular disorder Open
Inherited neuromuscular disorder (IND) is a broad-spectrum, clinically diverse group of diseases that are caused due to defects in the neurosystem, muscles and related tissue. Since IND may originate from mutations in hundreds of different…
View article: BlazeBVD: Make Scale-Time Equalization Great Again for Blind Video Deflickering
BlazeBVD: Make Scale-Time Equalization Great Again for Blind Video Deflickering Open
Developing blind video deflickering (BVD) algorithms to enhance video temporal consistency, is gaining importance amid the flourish of image processing and video generation. However, the intricate nature of video data complicates the train…
View article: Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis
Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis Open
Recent works in implicit representations, such as Neural Radiance Fields (NeRF), have advanced the generation of realistic and animatable head avatars from video sequences. These implicit methods are still confronted by visual artifacts an…
View article: Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory
Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory Open
This paper presents an analytical study of the oversmoothing issue in diffusion-based Graph Neural Networks (GNNs). Generalizing beyond extant approaches grounded in random walk analysis or particle systems, we approach this problem throug…
View article: Towards Optimal Adversarial Robust Q-learning with Bellman Infinity-error
Towards Optimal Adversarial Robust Q-learning with Bellman Infinity-error Open
Establishing robust policies is essential to counter attacks or disturbances affecting deep reinforcement learning (DRL) agents. Recent studies explore state-adversarial robustness and suggest the potential lack of an optimal robust policy…
View article: General Method for Solving Four Types of SAT Problems
General Method for Solving Four Types of SAT Problems Open
Existing methods provide varying algorithms for different types of Boolean satisfiability problems (SAT), lacking a general solution framework. Accordingly, this study proposes a unified framework DCSAT based on integer programming and rei…
View article: CARSS: Cooperative Attention-guided Reinforcement Subpath Synthesis for Solving Traveling Salesman Problem
CARSS: Cooperative Attention-guided Reinforcement Subpath Synthesis for Solving Traveling Salesman Problem Open
This paper introduces CARSS (Cooperative Attention-guided Reinforcement Subpath Synthesis), a novel approach to address the Traveling Salesman Problem (TSP) by leveraging cooperative Multi-Agent Reinforcement Learning (MARL). CARSS decompo…
View article: DiffBFR: Bootstrapping Diffusion Model for Blind Face Restoration
DiffBFR: Bootstrapping Diffusion Model for Blind Face Restoration Open
Blind face restoration (BFR) is important while challenging. Prior works prefer to exploit GAN-based frameworks to tackle this task due to the balance of quality and efficiency. However, these methods suffer from poor stability and adaptab…
View article: A-PSRO: A Unified Strategy Learning Method with Advantage Function for Normal-form Games
A-PSRO: A Unified Strategy Learning Method with Advantage Function for Normal-form Games Open
Solving Nash equilibrium is the key challenge in normal-form games with large strategy spaces, where open-ended learning frameworks offer an efficient approach. In this work, we propose an innovative unified open-ended learning framework A…
View article: Towards Consistent Video Editing with Text-to-Image Diffusion Models
Towards Consistent Video Editing with Text-to-Image Diffusion Models Open
Existing works have advanced Text-to-Image (TTI) diffusion models for video editing in a one-shot learning manner. Despite their low requirements of data and computation, these methods might produce results of unsatisfied consistency with …
View article: DiffBFR: Bootstrapping Diffusion Model Towards Blind Face Restoration
DiffBFR: Bootstrapping Diffusion Model Towards Blind Face Restoration Open
Blind face restoration (BFR) is important while challenging. Prior works prefer to exploit GAN-based frameworks to tackle this task due to the balance of quality and efficiency. However, these methods suffer from poor stability and adaptab…
View article: Exploring Over-smoothing in Graph Attention Networks from the Markov Chain Perspective
Exploring Over-smoothing in Graph Attention Networks from the Markov Chain Perspective Open
The over-smoothing problem causing the depth limitation is an obstacle of developing deep graph neural network (GNN). Compared with Graph Convolutional Networks (GCN), over-smoothing in Graph Attention Network (GAT) has not drawed enough a…
View article: Transforming Radiance Field with Lipschitz Network for Photorealistic 3D Scene Stylization
Transforming Radiance Field with Lipschitz Network for Photorealistic 3D Scene Stylization Open
Recent advances in 3D scene representation and novel view synthesis have witnessed the rise of Neural Radiance Fields (NeRFs). Nevertheless, it is not trivial to exploit NeRF for the photorealistic 3D scene stylization task, which aims to …
View article: StyO: Stylize Your Face in Only One-shot
StyO: Stylize Your Face in Only One-shot Open
This paper focuses on face stylization with a single artistic target. Existing works for this task often fail to retain the source content while achieving geometry variation. Here, we present a novel StyO model, ie. Stylize the face in onl…
View article: An Overview of Stochastic Quasi-Newton Methods for Large-Scale Machine Learning
An Overview of Stochastic Quasi-Newton Methods for Large-Scale Machine Learning Open
Numerous intriguing optimization problems arise as a result of the advancement of machine learning. The stochastic first-order method is the predominant choice for those problems due to its high efficiency. However, the negative effects of…
View article: Interpolation-aware models for train-test consistency in mixup
Interpolation-aware models for train-test consistency in mixup Open
Mixup is a learning principle that trains a neural network on convex combinations of pairs of examples and their labels. Despite of its good performance, there is an inherent inconsistency between training and testing in mixup, which makes…