Author name cluster

Feng Han

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.

10 papers

2 author rows

AIJ Journal 2026 Journal Article

Theoretical insights into CycleGAN: Analyzing approximation and estimation errors in unpaired data generation

Sun Luwei
Shen Dongrui
Feng Han

In this paper, we focus on analyzing the excess risk of the unpaired data generation model, called CycleGAN. Unlike classical GANs, CycleGAN not only transforms data between two unpaired distributions but also ensures the mappings are consistent, which is encouraged by the cycle-consistency term unique to CycleGAN. The increasing complexity of model structure and the addition of the cycle-consistency term in CycleGAN present new challenges for error analysis. By considering the impact of both the model architecture and training procedure, the risk is decomposed into two terms: approximation error and estimation error. These two error terms are analyzed separately and ultimately combined by considering the trade-off between them. Each component is rigorously analyzed; the approximation error through constructing approximations of the optimal transport maps, and the estimation error through establishing an upper bound using Rademacher complexity. Our analysis not only isolates these errors but also explores the trade-offs between them, which provides a theoretical insights of how CycleGAN's architecture and training procedures influence its performance.

Details DOI

IROS Conference 2025 Conference Paper

Active Training Data Selection for Gaussian Process-based Robot Dynamics Learning and Control

Feng Han
Yi Huang
Jingang Yi

Model-based robot control requires an accurate dynamics model and a machine learning-based method can extract robot dynamics from collected motion data by simulation and experiment. A Gaussian process (GP) has been used as one of the learning methods to obtain robot dynamics. To avoid large training datasets for learning robot dynamics, we propose an active training data selection strategy. The data sampling criteria are to minimize the probability density difference between the actual model and the GP-based estimate. Using such a criterion, the active training data strategy identifies where to sample the next data point for model training. We demonstrate the proposed active learning strategy with a 3-link robot arm in both fully actuated and underactuated modes. With the selected dataset containing 150 data points, the integrated probability density error compared with the entire dataset (over 30, 000 data points) is less than 0. 3. The experimental results confirm that the GP-based control performance is greater than that under the model-based control.

Details

AAAI Conference 2025 Conference Paper

DuMo: Dual Encoder Modulation Network for Precise Concept Erasure

Feng Han
Kai Chen
Chao Gong
Zhipeng Wei
Jingjing Chen
Yu-Gang Jiang

The exceptional generative capability of text-to-image models has raised substantial safety concerns regarding the generation of Not-Safe-For-Work (NSFW) content and potential copyright infringement. To address these concerns, previous methods safeguard the models by eliminating inappropriate concepts. Nonetheless, these models alter the parameters of the backbone network and exert considerable influences on the structural (low-frequency) components of the image, which undermines the model's ability to retain irrelevant concepts. In this work, we propose our Dual encoder Modulation network (DuMo), which achieves precise erasure of inappropriate target concepts with minimum impairment to non-target concepts. In contrast to previous methods, DuMo employs the Eraser with PRior Knowledge (EPR) module which modifies the skip connection features of the U-NET and primarily achieves concept erasure on details (high-frequency) components of the image. To minimize the demage to non-target concepts during erasure, the parameters of the backbone U-NET are frozen and the prior knowledge from the original skip connection features is introduced to the erasure process. Meanwhile, the phenomenon is observed that distinct erasing preferences for the image structure and details are demonstrated by the EPR at different timesteps and layers. Therefore, we adopt a novel Time-Layer MOdulation process (TLMO) that adjusts the erasure scale of EPR module's outputs across different layers and timesteps, automatically balancing the erasure effects and model's generative ability. Our method achieves state-of-the-art performance on Explicit Content Erasure (detecting only 34 nude parts), Cartoon Concept Removal (with an average LPIPS_da of 0.428, 0.113 higher than SOTA at 0.315), and Artistic Style Erasure (with an average LPIPS_da of 0.387, 0.088 higher than SOTA at 0.299), clearly outperforming alternative methods.

PDF Details DOI

ICRA Conference 2024 Conference Paper

Gaussian Process-Enhanced, External and Internal Convertible Form-Based Control of Underactuated Balance Robots

Feng Han
Jingang Yi

External and internal convertible (EIC) form-based motion control (i. e. , EIC-based control) is one of the effective approaches for underactuated balance robots. By sequentially controller design, trajectory tracking of the actuated subsystem and balance of the unactuated subsystem can be achieved simultaneously. However, with certain conditions, there exists uncontrolled robot motion under the EIC-based control. We first identify these conditions and then propose an enhanced EIC-based control with a Gaussian process data-driven robot dynamic model. Under the new enhanced EIC-based control, the stability and performance of the closed-loop system are guaranteed. We demonstrate the GP-enhanced control experimentally using two examples of underactuated balance robots.

Details

YNIMG Journal 2024 Journal Article

Sex-specific age-related differences in cerebrospinal fluid clearance assessed by resting-state functional magnetic resonance imaging

Feng Han
Xufu Liu
Yifan Yang
Xiao Liu

Cerebrospinal fluid (CSF) flow may assist the clearance of brain wastes, such as amyloid-β (Aβ) and tau, and thus play an important role in aging and dementias. However, a lack of non-invasive tools to assess the CSF dynamics-related clearance in humans hindered the understanding of the relevant changes in healthy aging. The global infra-slow (<0.1 Hz) brain activity measured by the global mean resting-state fMRI signal (gBOLD) was recently found to be coupled by large CSF movements. This coupling has been found to correlate with various pathologies of Alzheimer's disease (AD), particularly Aβ pathology, linking it to waste clearance. Using resting-state fMRI data from a group of 719 healthy aging participants, we examined the sex-specific differences of the gBOLD-CSF coupling over a wide age range between 36-100 years of age. We found that this coupling index remains stable before around age 55 and then starts to decline afterward, particularly in females. Menopause may contribute to the accelerated decline in females.

Details DOI

ICRA Conference 2023 Conference Paper

On the Learned Balance Manifold of Underactuated Balance Robots

Feng Han
Jingang Yi

Tracking control of underactuated balance robots needs to estimate balance profiles, that is, balance equilibrium manifold (BEM) of the unactuated subsystems. We present a learning-based approach to obtain the balance manifold for underactuated balance robots. We first establish the relationship between the BEM and the zero dynamics of the underactuated balance robots. The analysis shows that the BEM is a close approximation of the equilibria of the zero dynamics under perfectly tracking control. A Gaussian process learning-based method is proposed to estimate and obtain the BEM and zero dynamics, avoiding the direct inversion of the physics-based robot dynamic model. We demonstrate the analysis and applications experimentally on a rotary inverted pendulum and a bipedal robot.

Details

YNIMG Journal 2022 Journal Article

An orderly sequence of autonomic and neural events at transient arousal changes

Yameng Gu
Feng Han
Lucas E. Sainburg
Margeaux M. Schade
Orfeu M. Buxton
Jeff H. Duyn
Xiao Liu

Resting-state functional magnetic resonance imaging (rsfMRI) allows the study of functional brain connectivity based on spatially structured variations in neuronal activity. Proper evaluation of connectivity requires removal of non-neural contributions to the fMRI signal, in particular hemodynamic changes associated with autonomic variability. Regression analysis based on autonomic indicator signals has been used for this purpose, but may be inadequate if neuronal and autonomic activities covary. To investigate this potential co-variation, we performed rsfMRI experiments while concurrently acquiring electroencephalography (EEG) and autonomic indicator signals, including heart rate, respiratory depth, and peripheral vascular tone. We identified a recurrent and systematic spatiotemporal pattern of fMRI (named as fMRI cascade), which features brief signal reductions in salience and default-mode networks and the thalamus, followed by a biphasic global change with a sensory-motor dominance. This fMRI cascade, which was mostly observed during eyes-closed condition, was accompanied by large EEG and autonomic changes indicative of arousal modulations. Importantly, the removal of the fMRI cascade dynamics from rsfMRI diminished its correlations with various signals. These results suggest that the rsfMRI correlations with various physiological and neural signals are not independent but arise, at least partly, from the fMRI cascades and associated neural and physiological changes at arousal modulations.

Details DOI

JAIR Journal 2022 Journal Article

C-Face: Using Compare Face on Face Hallucination for Low-Resolution Face Recognition

Feng Han
Xudong Wang
Furao Shen
Jian Zhao

Face hallucination is a task of generating high-resolution (HR) face images from low-resolution (LR) inputs, which is a subfield of the general image super-resolution. However, most of the previous methods only consider the visual effect, ignoring how to maintain the identity of the face. In this work, we propose a novel face hallucination model, called C-Face network, which can generate HR images with high visual quality while preserving the identity information. A face recognition network is used to extract the identity features in the training process. In order to make the reconstructed face images keep the identity information to a great extent, a novel metric, i.e., C-Face loss, is proposed. We also propose a new training algorithm to deal with the convergence problem. Moreover, since our work mainly focuses on the recognition accuracy of the output, we integrate face recognition into the face hallucination process which ensures that the model can be used in real scenarios. Extensive experiments on two large scale face datasets demonstrate that our C-Face network has the best performance compared with other state-of-the-art methods.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Scaling Multimodal Pre-Training via Cross-Modality Gradient Harmonization

Junru Wu
Yi Liang
Feng Han
Hassan Akbari
Zhangyang Wang
Cong Yu

Self-supervised pre-training recently demonstrates success on large-scale multimodal data, and state-of-the-art contrastive learning methods often enforce the feature consistency from cross-modality inputs, such as video/audio or video/text pairs. Despite its convenience to formulate and leverage in practice, such cross-modality alignment (CMA) is only a weak and noisy supervision, since two modalities can be semantically misaligned even they are temporally aligned. For example, even in the (often adopted) instructional videos, a speaker can sometimes refer to something that is not visually present in the current frame; and the semantic misalignment would only be more unpredictable for the raw videos collected from unconstrained internet sources. We conjecture that might cause conflicts and biases among modalities, and may hence prohibit CMA from scaling up to training with larger and more heterogeneous data. This paper first verifies our conjecture by observing that, even in the latest VATT pre-training using only narrated videos, there exist strong gradient conflicts between different CMA losses within the same sample triplet (video, audio, text), indicating them as the noisy source of supervision. We then propose to harmonize such gradients during pre-training, via two techniques: (i) cross-modality gradient realignment: modifying different CMA loss gradients for one sample triplet, so that their gradient directions are in more agreement; and (ii) gradient-based curriculum learning: leveraging the gradient conflict information on an indicator of sample noisiness, to develop a curriculum learning strategy to prioritize training with less noisy sample triplets. Applying those gradient harmonization techniques to pre-training VATT on the HowTo100M dataset, we consistently improve its performance on different downstream tasks. Moreover, we are able to scale VATT pre-training to more complicated non-narrative Youtube8M dataset to further improve the state-of-the-arts.

PDF Details

YNIMG Journal 2020 Journal Article

Neuroimaging contrast across the cortical hierarchy is the feature maximally linked to behavior and demographics

Feng Han
Yameng Gu
Gregory L. Brown
Xiang Zhang
Xiao Liu

An essential task of neuroscience is to elucidate the relationship between brain activity, brain structure, and human behavior. This study aims to understand this 3-way relationship by studying the population covariance of resting-state functional connectivity, cortical thickness, and behavioral/demographic measures in a large cohort of individuals. Using a data-driven canonical correlation analysis, we found that maximal pairwise correlations between the three modalities are approximately along the same direction across subjects, which is characterized by the change of the overall positive-negative trait of human behavior. More importantly, this behavioral change is associated with a divergent modulation of both resting-state connectivity and cortical thickness across cortical hierarchies between the higher-order cognitive networks and lower-order sensory/motor regions. The findings suggest that the cross-hierarchy contrast of structural and functional brain measures is tightly linked to the overall positive-negative trait of human behavior/demographics.

Details DOI