arXiv subjects

Cheng Chen

Publications and source records attributed to Cheng Chen.

At least 217 records · Page 12Linked to original sources

arXiv2024

Adapting CSI-Guided Imaging Across Diverse Environments: An Experimental Study Leveraging Continuous Learning

This study explores the feasibility of adapting CSI-guided imaging across varied environments. Focusing on continuous model learning through continuous updates, we investigate CSI-Imager's adaptability in dynamically changing settings, specifically transitioning from an office to an industrial environment. Unlike traditional approaches that may require retraining for new environments, our experimental study aims to validate the potential of CSI-guided imaging to maintain accurate imaging performance through Continuous Learning (CL). By conducting experiments across different scenarios and settings, this work contributes to understanding the limitations and capabilities of existing CSI-guided imaging systems in adapting to new environmental contexts.

eess.IV↗

arXiv2024

Radiology-GPT: A Large Language Model for Radiology

We introduce Radiology-GPT, a large language model for radiology. Using an instruction tuning approach on an extensive dataset of radiology domain knowledge, Radiology-GPT demonstrates superior performance compared to general language models such as StableLM, Dolly and LLaMA. It exhibits significant versatility in radiological diagnosis, research, and communication. This work serves as a catalyst for future developments in clinical NLP. The successful implementation of Radiology-GPT is indicative of the potential of localizing generative large language models, specifically tailored for distinctive medical specialties, while ensuring adherence to privacy standards such as HIPAA. The prospect of developing individualized, large-scale language models that cater to specific needs of various hospitals presents a promising direction. The fusion of conversational competence and domain-specific knowledge in these models is set to foster future development in healthcare AI. A demo of Radiology-GPT is available at https://huggingface.co/spaces/allen-eric/radiology-gpt.

cs.CL↗

arXiv2024

Sculpt3D: Multi-View Consistent Text-to-3D Generation with Sparse 3D Prior

Recent works on text-to-3d generation show that using only 2D diffusion supervision for 3D generation tends to produce results with inconsistent appearances (e.g., faces on the back view) and inaccurate shapes (e.g., animals with extra legs). Existing methods mainly address this issue by retraining diffusion models with images rendered from 3D data to ensure multi-view consistency while struggling to balance 2D generation quality with 3D consistency. In this paper, we present a new framework Sculpt3D that equips the current pipeline with explicit injection of 3D priors from retrieved reference objects without re-training the 2D diffusion model. Specifically, we demonstrate that high-quality and diverse 3D geometry can be guaranteed by keypoints supervision through a sparse ray sampling approach. Moreover, to ensure accurate appearances of different views, we further modulate the output of the 2D diffusion model to the correct patterns of the template views without altering the generated object's style. These two decoupled designs effectively harness 3D information from reference objects to generate 3D objects while preserving the generation quality of the 2D diffusion model. Extensive experiments show our method can largely improve the multi-view consistency while retaining fidelity and diversity. Our project page is available at: https://stellarcheng.github.io/Sculpt3D/.

cs.CV↗

arXiv2024

Automating Catheterization Labs with Real-Time Perception

For decades, three-dimensional C-arm Cone-Beam Computed Tomography (CBCT) imaging system has been a critical component for complex vascular and nonvascular interventional procedures. While it can significantly improve multiplanar soft tissue imaging and provide pre-treatment target lesion roadmapping and guidance, the traditional workflow can be cumbersome and time-consuming, especially for less experienced users. To streamline this process and enhance procedural efficiency overall, we proposed a visual perception system, namely AutoCBCT, seamlessly integrated with an angiography suite. This system dynamically models both the patient's body and the surgical environment in real-time. AutoCBCT enables a novel workflow with automated positioning, navigation and simulated test-runs, eliminating the need for manual operations and interactions. The proposed system has been successfully deployed and studied in both lab and clinical settings, demonstrating significantly improved workflow efficiency.

cs.CV↗

arXiv2024

Orbital torque switching in perpendicularly magnetized materials

The orbital Hall effect in light materials has attracted considerable attention for developing novel orbitronic devices. Here we investigate the orbital torque efficiency and demonstrate the switching of the perpendicularly magnetized materials through the orbital Hall material (OHM), i.e., Zirconium (Zr). The orbital torque efficiency of approximately 0.78 is achieved in the Zr OHM with the perpendicularly magnetized [Co/Pt]3 sample, which significantly surpasses that of the perpendicularly magnetized CoFeB/Gd/CoFeB sample (approximately 0.04). Such notable difference is attributed to the different spin-orbit correlation strength between the [Co/Pt]3 sample and the CoFeB/Gd/CoFeB sample, which has been confirmed through the theoretical calculations. Furthermore, the full magnetization switching of the [Co/Pt]3 sample with a switching current density of approximately 2.6x106 A/cm2 has been realized through Zr, which even outperforms that of the W spin Hall material. Our finding provides a guideline to understand orbital torque efficiency and paves the way to develop energy-efficient orbitronic devices.

cond-mat.mtrl-sci↗

arXiv2024

Assisted Learning for Organizations with Limited Imbalanced Data

In the era of big data, many big organizations are integrating machine learning into their work pipelines to facilitate data analysis. However, the performance of their trained models is often restricted by limited and imbalanced data available to them. In this work, we develop an assisted learning framework for assisting organizations to improve their learning performance. The organizations have sufficient computation resources but are subject to stringent data-sharing and collaboration policies. Their limited imbalanced data often cause biased inference and sub-optimal decision-making. In assisted learning, an organizational learner purchases assistance service from an external service provider and aims to enhance its model performance within only a few assistance rounds. We develop effective stochastic training algorithms for both assisted deep learning and assisted reinforcement learning. Different from existing distributed algorithms that need to frequently transmit gradients or models, our framework allows the learner to only occasionally share information with the service provider, but still obtain a model that achieves near-oracle performance as if all the data were centralized.

cs.LG↗

arXiv2024

CSI-Inpainter: Enabling Visual Scene Recovery from CSI Time Sequences for Occlusion Removal

Introducing CSI-Inpainter, a pioneering approach for occlusion removal using Channel State Information (CSI) time sequences, this work propels the application of wireless signal processing into the realm of visual scene recovery. Departing from traditional occlusion removal, CSI-Inpainter leverages CSI data to construct and refine obscured visual elements in a scene, facilitating recovery independent of lighting conditions. Validated through comprehensive testing in both office and industrial environments, CSI-Inpainter demonstrates a robust capacity for discerning and reconstructing occluded segments, establishing a new frontier for obstacle removal. This first-of-its-kind framework offers a transformative perspective on how environmental visual information can be extracted from CSI, thereby broadening the scope for computer vision applications in everyday contexts.

eess.IV↗

arXiv2024

Understanding oscillating features of the timelike nucleon electromagnetic form factors within the extending vector meson dominance model

We investigate the nonmonotonic behavior observed in the time-like nucleon electromagnetic form factors. Using a phenomenological extending vector meson dominance model, where the ground states $ρ$ and $ω$ and their excited states $ρ(2D)$, $ω(3D)$, and $ω(5S)$ are taken into account, we have successfully reproduced the cross sections of $e^+ e^- \to p \bar{p}$ and $e^+ e^- \to n \bar{n}$ reactions. Furthermore, we have derived the nucleon electromagnetic form factors in the time-like region, and it is found that the so-called periodic behaviour of the nucleon effective form factors is not confirmed. However, there are indeed nonmonotonic structures in the line shape of nucleon effective form factors, which can be naturally reproduced by considering the contributions from the low-lying excited vector states.

nucl-th↗

arXiv2024

$e^+ e^- \to Λ^+_c \barΛ^-_c$ cross sections and the $Λ_c^+$ electromagnetic form factors within the extended vector meson dominance model

Within the extended vector meson dominance model, we investigate the $e^+ e^- \to Λ^+_c \barΛ^-_c$ reaction and the electromagnetic form factors of the charmed baryon $Λ_c^+$. The model parameters are determined by fitting them to the cross sections of the process $e^+e^-\rightarrow Λ_c^+ \barΛ_c^-$ and the magnetic form factor $|G_M|$ of $Λ^+_c$. By considering four charmoniumlike states, called $ψ(4500)$, $ψ(4660)$, $ψ(4790)$, and $ψ(4900)$, we can well describe the current data on the $e^+ e^- \to Λ^+_c \barΛ^-_c$ reaction from the reaction threshold up to $4.96 \ \mathrm{GeV}$. In addition to the total cross sections and $|G_M|$, the ratio $|G_E/G_M|$ and effective form factor $|G_{\mathrm{eff}}|$ for $Λ^+_c$ are also calculated, and found that these calculations are consistent with the experimental data. Within the fitted model parameters, we have also estimated the charge radius of the charmed $Λ_c^+$ baryon.

hep-ph↗

arXiv2024

Enhancing a Many-body Dipolar Rydberg Tweezer Array with Arbitrary Local Controls

We implement and characterize a protocol that enables arbitrary local controls in a dipolar atom array, where the degree of freedom is encoded in a pair of Rydberg states. Our approach relies on a combination of local addressing beams and global microwave fields. Using this method, we directly prepare two different types of three-atom entangled states, including a W-state and a state exhibiting finite chirality. We verify the nature of the underlying entanglement by performing quantum state tomography. Finally, leveraging our ability to measure multi-basis, multi-body observables, we explore the adiabatic preparation of low-energy states in a frustrated geometry consisting of a pair of triangular plaquettes. By using local addressing to tune the symmetry of the initial state, we demonstrate the ability to prepare correlated states distinguished only by correlations of their chirality (a fundamentally six-body observable). Our protocol is generic, allowing for rotations on arbitrary subgroups of atoms within the array at arbitrary times during the experiment; this extends the scope of capabilities for quantum simulations of the dipolar XY model.

quant-ph↗

arXiv2024

Producing Plankton Classifiers that are Robust to Dataset Shift

Modern plankton high-throughput monitoring relies on deep learning classifiers for species recognition in water ecosystems. Despite satisfactory nominal performances, a significant challenge arises from Dataset Shift, which causes performances to drop during deployment. In our study, we integrate the ZooLake dataset with manually-annotated images from 10 independent days of deployment, serving as test cells to benchmark Out-Of-Dataset (OOD) performances. Our analysis reveals instances where classifiers, initially performing well in In-Dataset conditions, encounter notable failures in practical scenarios. For example, a MobileNet with a 92% nominal test accuracy shows a 77% OOD accuracy. We systematically investigate conditions leading to OOD performance drops and propose a preemptive assessment method to identify potential pitfalls when classifying new data, and pinpoint features in OOD images that adversely impact classification. We present a three-step pipeline: (i) identifying OOD degradation compared to nominal test performance, (ii) conducting a diagnostic analysis of degradation causes, and (iii) providing solutions. We find that ensembles of BEiT vision transformers, with targeted augmentations addressing OOD robustness, geometric ensembling, and rotation-based test-time augmentation, constitute the most robust model, which we call BEsT model. It achieves an 83% OOD accuracy, with errors concentrated on container classes. Moreover, it exhibits lower sensitivity to dataset shift, and reproduces well the plankton abundances. Our proposed pipeline is applicable to generic plankton classifiers, contingent on the availability of suitable test cells. By identifying critical shortcomings and offering practical procedures to fortify models against dataset shift, our study contributes to the development of more reliable plankton classification technologies.

cs.CV↗

arXiv2024

Fast Updating Truncated SVD for Representation Learning with Sparse Matrices

Updating a truncated Singular Value Decomposition (SVD) is crucial in representation learning, especially when dealing with large-scale data matrices that continuously evolve in practical scenarios. Aligning SVD-based models with fast-paced updates becomes increasingly important. Existing methods for updating truncated SVDs employ Rayleigh-Ritz projection procedures, where projection matrices are augmented based on original singular vectors. However, these methods suffer from inefficiency due to the densification of the update matrix and the application of the projection to all singular vectors. To address these limitations, we introduce a novel method for dynamically approximating the truncated SVD of a sparse and temporally evolving matrix. Our approach leverages sparsity in the orthogonalization process of augmented matrices and utilizes an extended decomposition to independently store projections in the column space of singular vectors. Numerical experiments demonstrate a remarkable efficiency improvement of an order of magnitude compared to previous methods. Remarkably, this improvement is achieved while maintaining a comparable precision to existing approaches.

math.NA↗

arXiv2024

Towards Redundancy-Free Sub-networks in Continual Learning

Catastrophic Forgetting (CF) is a prominent issue in continual learning. Parameter isolation addresses this challenge by masking a sub-network for each task to mitigate interference with old tasks. However, these sub-networks are constructed relying on weight magnitude, which does not necessarily correspond to the importance of weights, resulting in maintaining unimportant weights and constructing redundant sub-networks. To overcome this limitation, inspired by information bottleneck, which removes redundancy between adjacent network layers, we propose \textbf{\underline{I}nformation \underline{B}ottleneck \underline{M}asked sub-network (IBM)} to eliminate redundancy within sub-networks. Specifically, IBM accumulates valuable information into essential weights to construct redundancy-free sub-networks, not only effectively mitigating CF by freezing the sub-networks but also facilitating new tasks training through the transfer of valuable knowledge. Additionally, IBM decomposes hidden representations to automate the construction process and make it flexible. Extensive experiments demonstrate that IBM consistently outperforms state-of-the-art methods. Notably, IBM surpasses the state-of-the-art parameter isolation method with a 70\% reduction in the number of parameters within sub-networks and an 80\% decrease in training time.

cs.LG↗

arXiv2024

A Comprehensive Review of Pre-Darcy Flows in Low-Permeability Porous Media

This paper reviews theories, experimental data, and modeling methods for pre-Darcy flow in low-permeability porous media, where Darcy velocity shows nonlinear dependence on pressure gradients at sufficiently low pressures, a deviation from Darcy's law. It begins by explaining the fundamental mechanisms of pre-Darcy flow, focusing on its unique characteristics like non-linear pressure gradients and fluid-rock interactions. Next, the paper compiles experimental studies on low-permeability geomaterials such as tight sandstones, shales, and clays, detailing methodologies employed, including core sample preparation, permeability measurement techniques, and threshold pressure gradient assessments. The experiments' findings, showing how pore geometry, fluid type, and pressure conditions affect pre-Darcy flow onset, are discussed. The review then covers empirical and theoretical models, plus simulation methods developed for interpreting data on pre-Darcy flow. It concludes by highlighting challenges in conducting and interpreting these experiments, suggesting directions for future research. This comprehensive analysis aims to assist those studying fluid dynamics in low-permeability geomaterials and has implications for applications like shale oil and gas recovery, contaminant transport in low-permeability aquifers, and geological nuclear waste disposal.

physics.geo-ph↗

arXiv2023

Robustness Verification of Deep Reinforcement Learning Based Control Systems using Reward Martingales

Deep Reinforcement Learning (DRL) has gained prominence as an effective approach for control systems. However, its practical deployment is impeded by state perturbations that can severely impact system performance. Addressing this critical challenge requires robustness verification about system performance, which involves tackling two quantitative questions: (i) how to establish guaranteed bounds for expected cumulative rewards, and (ii) how to determine tail bounds for cumulative rewards. In this work, we present the first approach for robustness verification of DRL-based control systems by introducing reward martingales, which offer a rigorous mathematical foundation to characterize the impact of state perturbations on system performance in terms of cumulative rewards. Our verified results provide provably quantitative certificates for the two questions. We then show that reward martingales can be implemented and trained via neural networks, against different types of control policies. Experimental results demonstrate that our certified bounds tightly enclose simulation outcomes on various DRL-based control systems, indicating the effectiveness and generality of the proposed approach.

cs.AI↗

arXiv2023

Learn to Optimize Denoising Scores for 3D Generation: A Unified and Improved Diffusion Prior on NeRF and 3D Gaussian Splatting

We propose a unified framework aimed at enhancing the diffusion priors for 3D generation tasks. Despite the critical importance of these tasks, existing methodologies often struggle to generate high-caliber results. We begin by examining the inherent limitations in previous diffusion priors. We identify a divergence between the diffusion priors and the training procedures of diffusion models that substantially impairs the quality of 3D generation. To address this issue, we propose a novel, unified framework that iteratively optimizes both the 3D model and the diffusion prior. Leveraging the different learnable parameters of the diffusion prior, our approach offers multiple configurations, affording various trade-offs between performance and implementation complexity. Notably, our experimental results demonstrate that our method markedly surpasses existing techniques, establishing new state-of-the-art in the realm of text-to-3D generation. Furthermore, our approach exhibits impressive performance on both NeRF and the newly introduced 3D Gaussian Splatting backbones. Additionally, our framework yields insightful contributions to the understanding of recent score distillation methods, such as the VSD and DDS loss.

cs.CV↗

arXiv2023

Divide-and-Conquer Strategy for Large-Scale Dynamic Bayesian Network Structure Learning

Dynamic Bayesian Networks (DBNs), renowned for their interpretability, have become increasingly vital in representing complex stochastic processes in various domains such as gene expression analysis, healthcare, and traffic prediction. Structure learning of DBNs from data is challenging, particularly for datasets with thousands of variables. Most current algorithms for DBN structure learning are adaptations from those used in static Bayesian Networks (BNs), and are typically focused on small-scale problems. In order to solve large-scale problems while taking full advantage of existing algorithms, this paper introduces a novel divide-and-conquer strategy, originally developed for static BNs, and adapts it for large-scale DBN structure learning. In this work, we specifically concentrate on 2 Time-sliced Bayesian Networks (2-TBNs), a special class of DBNs. Furthermore, we leverage the prior knowledge of 2-TBNs to enhance the performance of the strategy we introduce. Our approach significantly improves the scalability and accuracy of 2-TBN structure learning. Experimental results demonstrate the effectiveness of our method, showing substantial improvements over existing algorithms in both computational efficiency and structure learning accuracy. On problem instances with more than 1,000 variables, our approach improves two accuracy metrics by 74.45% and 110.94% on average , respectively, while reducing runtime by 93.65% on average.

cs.LG↗

arXiv2023

Continual Referring Expression Comprehension via Dual Modular Memorization

Referring Expression Comprehension (REC) aims to localize an image region of a given object described by a natural-language expression. While promising performance has been demonstrated, existing REC algorithms make a strong assumption that training data feeding into a model are given upfront, which degrades its practicality for real-world scenarios. In this paper, we propose Continual Referring Expression Comprehension (CREC), a new setting for REC, where a model is learning on a stream of incoming tasks. In order to continuously improve the model on sequential tasks without forgetting prior learned knowledge and without repeatedly re-training from a scratch, we propose an effective baseline method named Dual Modular Memorization (DMM), which alleviates the problem of catastrophic forgetting by two memorization modules: Implicit-Memory and Explicit-Memory. Specifically, the former module aims to constrain drastic changes to important parameters learned on old tasks when learning a new task; while the latter module maintains a buffer pool to dynamically select and store representative samples of each seen task for future rehearsal. We create three benchmarks for the new CREC setting, by respectively re-splitting three widely-used REC datasets RefCOCO, RefCOCO+ and RefCOCOg into sequential tasks. Extensive experiments on the constructed benchmarks demonstrate that our DMM method significantly outperforms other alternatives, based on two popular REC backbones. We make the source code and benchmarks publicly available to foster future progress in this field: https://github.com/zackschen/DMM.

cs.CV↗