arXiv subjects

Chao Tang

Publications and source records attributed to Chao Tang.

At least 55 records · Page 3Linked to original sources

arXiv2024

Commonsense Scene Graph-based Target Localization for Object Search

Object search is a fundamental skill for household robots, yet the core problem lies in the robot's ability to locate the target object accurately. The dynamic nature of household environments, characterized by the arbitrary placement of daily objects by users, makes it challenging to perform target localization. To efficiently locate the target object, the robot needs to be equipped with knowledge at both the object and room level. However, existing approaches rely solely on one type of knowledge, leading to unsatisfactory object localization performance and, consequently, inefficient object search processes. To address this problem, we propose a commonsense scene graph-based target localization, CSG-TL, to enhance target object search in the household environment. Given the pre-built map with stationary items, the robot models the room-level knowledge with object-level commonsense knowledge generated by a large language model (LLM) to a commonsense scene graph (CSG), supporting both types of knowledge for CSG-TL. To demonstrate the superiority of CSG-TL on target localization, extensive experiments are performed on the real-world ScanNet dataset and the AI2THOR simulator. Moreover, we have extended CSG-TL to an object search framework, CSG-OS, validated in both simulated and real-world environments. Code and videos are available at https://sites.google.com/view/csg-os.

cs.RO↗

arXiv2024

RTAGrasp: Learning Task-Oriented Grasping from Human Videos via Retrieval, Transfer, and Alignment

Task-oriented grasping (TOG) is crucial for robots to accomplish manipulation tasks, requiring the determination of TOG positions and directions. Existing methods either rely on costly manual TOG annotations or only extract coarse grasping positions or regions from human demonstrations, limiting their practicality in real-world applications. To address these limitations, we introduce RTAGrasp, a Retrieval, Transfer, and Alignment framework inspired by human grasping strategies. Specifically, our approach first effortlessly constructs a robot memory from human grasping demonstration videos, extracting both TOG position and direction constraints. Then, given a task instruction and a visual observation of the target object, RTAGrasp retrieves the most similar human grasping experience from its memory and leverages semantic matching capabilities of vision foundation models to transfer the TOG constraints to the target object in a training-free manner. Finally, RTAGrasp aligns the transferred TOG constraints with the robot's action for execution. Evaluations on the public TOG benchmark, TaskGrasp dataset, show the competitive performance of RTAGrasp on both seen and unseen object categories compared to existing baseline methods. Real-world experiments further validate its effectiveness on a robotic arm. Our code, appendix, and video are available at \url{https://sites.google.com/view/rtagrasp/home}.

cs.RO↗

arXiv2024

The AIRI plug-and-play algorithm for image reconstruction in radio-interferometry: variations and robustness

Plug-and-Play (PnP) algorithms are appealing alternatives to proximal algorithms when solving inverse imaging problems. By learning a Deep Neural Network (DNN) denoiser behaving as a proximal operator, one waives the computational complexity of optimisation algorithms induced by sophisticated image priors, and the sub-optimality of handcrafted priors compared to DNNs. Such features are highly desirable in radio-interferometric (RI) imaging, where precision and scalability of the image reconstruction process are key. In previous work, we introduced AIRI, PnP counterpart to the unconstrained variant of the SARA optimisation algorithm, relying on a forward-backward algorithmic backbone. Here, we introduce variations of AIRI towards a more general and robust PnP paradigm in RI imaging. Firstly, we show that the AIRI denoisers can be used without any alteration to instantiate a PnP counterpart to the constrained SARA optimisation algorithm itself, relying on a primal-dual forward-backward algorithmic backbone, thus extending the remit of the AIRI paradigm. Secondly, we show that AIRI algorithms are robust to strong variations in the nature of the training dataset, with denoisers trained on medical images yielding similar reconstruction quality to those trained on astronomical images. Thirdly, we develop a functionality to quantify the model uncertainty introduced by the randomness in the training process. We validate the image reconstruction and uncertainty quantification functionality of AIRI algorithms against the SARA family and CLEAN, both in simulation and on real data of the ESO 137-006 galaxy acquired with the MeerKAT telescope. AIRI code is available in the BASPLib code library on GitHub.

eess.IV↗

arXiv2024

Scene-wise Adaptive Network for Dynamic Cold-start Scenes Optimization in CTR Prediction

In the realm of modern mobile E-commerce, providing users with nearby commercial service recommendations through location-based online services has become increasingly vital. While machine learning approaches have shown promise in multi-scene recommendation, existing methodologies often struggle to address cold-start problems in unprecedented scenes: the increasing diversity of commercial choices, along with the short online lifespan of scenes, give rise to the complexity of effective recommendations in online and dynamic scenes. In this work, we propose Scene-wise Adaptive Network (SwAN), a novel approach that emphasizes high-performance cold-start online recommendations for new scenes. Our approach introduces several crucial capabilities, including scene similarity learning, user-specific scene transition cognition, scene-specific information construction for the new scene, and enhancing the diverged logical information between scenes. We demonstrate SwAN's potential to optimize dynamic multi-scene recommendation problems by effectively online handling cold-start recommendations for any newly arrived scenes. More encouragingly, SwAN has been successfully deployed in Meituan's online catering recommendation service, which serves millions of customers per day, and SwAN has achieved a 5.64% CTR index improvement relative to the baselines and a 5.19% increase in daily order volume proportion.

cs.IR↗

arXiv2024

Estimating before Debiasing: A Bayesian Approach to Detaching Prior Bias in Federated Semi-Supervised Learning

Federated Semi-Supervised Learning (FSSL) leverages both labeled and unlabeled data on clients to collaboratively train a model.In FSSL, the heterogeneous data can introduce prediction bias into the model, causing the model's prediction to skew towards some certain classes. Existing FSSL methods primarily tackle this issue by enhancing consistency in model parameters or outputs. However, as the models themselves are biased, merely constraining their consistency is not sufficient to alleviate prediction bias. In this paper, we explore this bias from a Bayesian perspective and demonstrate that it principally originates from label prior bias within the training data. Building upon this insight, we propose a debiasing method for FSSL named FedDB. FedDB utilizes the Average Prediction Probability of Unlabeled Data (APP-U) to approximate the biased prior.During local training, FedDB employs APP-U to refine pseudo-labeling through Bayes' theorem, thereby significantly reducing the label prior bias. Concurrently, during the model aggregation, FedDB uses APP-U from participating clients to formulate unbiased aggregate weights, thereby effectively diminishing bias in the global model. Experimental results show that FedDB can surpass existing FSSL methods. The code is available at https://github.com/GuogangZhu/FedDB.

cs.LG↗

arXiv2024

Scalable Non-Cartesian Magnetic Resonance Imaging with R2D2

We propose a new approach for non-Cartesian magnetic resonance image reconstruction. While unrolled architectures provide robustness via data-consistency layers, embedding measurement operators in Deep Neural Network (DNN) can become impractical at large scale. Alternative Plug-and-Play (PnP) approaches, where the denoising DNNs are blind to the measurement setting, are not affected by this limitation and have also proven effective, but their highly iterative nature also affects scalability. To address this scalability challenge, we leverage the "Residual-to-Residual DNN series for high-Dynamic range imaging (R2D2)" approach recently introduced in astronomical imaging. R2D2's reconstruction is formed as a series of residual images, iteratively estimated as outputs of DNNs taking the previous iteration's image estimate and associated data residual as inputs. The method can be interpreted as a learned version of the Matching Pursuit algorithm. We demonstrate R2D2 in simulation, considering radial k-space sampling acquisition sequences. Our preliminary results suggest that R2D2 achieves: (i) suboptimal performance compared to its unrolled incarnation R2D2-Net, which is however non-scalable due to the necessary embedding of NUFFT-based data-consistency layers; (ii) superior reconstruction quality to a scalable version of R2D2-Net embedding an FFT-based approximation for data consistency; (iii) superior reconstruction quality to PnP, while only requiring few iterations.

eess.IV↗

arXiv2024

An Invertible All-optical Logic Gate on Chip

We demonstrate an invertible all-optical gate on chip, with the roles of control and signal switchable by slightly adjusting their relative arrival time at the gate. It is based on quantum Zeno blockade driven by sum-frequency generation in a periodic-poled lithium niobate microring resonator. For two nearly-identical nanosecond pulses, the later arriving pulse is modulated by the earlier arriving one, resulting in 2.4 and 3.9 power extinction between the two, respectively, when their peak power is 1 mW and 2 mW. Our results, while to be improved and enriched, herald a new paradigm of logical gates and circuits for exotic applications.

physics.optics↗

arXiv2024

Parametric All-Optical Modulation on Chip

We demonstrate parametric all-optical modulation in a periodically-poled lithium niobate microring resonator on chip. It employs quantum Zeno blockade between two distinct waves, a signal and a pump, through their sum-frequency generation at a large per-photon efficiency of 8.2 MHz. With nanosecond pump pulses at 6 mW peak power, 85.7% modulation extinction is observed, marking over 30~times efficiency improvement across various previous implementations. With only 2 mW pump peak power, 43.0% modulation extinction is observed for a doubly-stronger signal at 4 mW. This demonstrates, for the first time, that optical transistors with cascadability and fan-out are possible with just parametric nonlinear optics. These results, together with inherent advantages in such photonic integrated circuits, open the door to scalable technology for all-optical and quantum information processing.

physics.optics↗

arXiv2023

HKTGNN: Hierarchical Knowledge Transferable Graph Neural Network-based Supply Chain Risk Assessment

The strength of a supply chain is an important measure of a country's or region's technical advancement and overall competitiveness. Establishing supply chain risk assessment models for effective management and mitigation of potential risks has become increasingly crucial. As the number of businesses grows, the important relationships become more complicated and difficult to measure. This emphasizes the need of extracting relevant information from graph data. Previously, academics mostly employed knowledge inference to increase the visibility of links between nodes in the supply chain. However, they have not solved the data hunger problem of single node feature characteristics. We propose a hierarchical knowledge transferable graph neural network-based (HKTGNN) supply chain risk assessment model to address these issues. Our approach is based on current graph embedding methods for assessing corporate investment risk assessment. We embed the supply chain network corresponding to individual goods in the supply chain using the graph embedding module, resulting in a directed homogeneous graph with just product nodes. This reduces the complicated supply chain network into a basic product network. It addresses difficulties using the domain difference knowledge transferable module based on centrality, which is presented by the premise that supply chain feature characteristics may be biased in the actual world. Meanwhile, the feature complement and message passing will alleviate the data hunger problem, which is driven by domain differences. Our model outperforms in experiments on a real-world supply chain dataset. We will give an equation to prove that our comparative experiment is both effective and fair.

cs.LG↗

arXiv2023

A Highly Efficient and Pure Few-Photon Source on Chip

We report on multi-photon statistics of correlated twin beams produced in a periodic poled micro-ring resonator on thin-film lithium niobate. Owing to high cavity confinement and near perfect quasi-phase matching, the photons pairs are produced efficiently in single modes at rates reaching 27 MHz per $μ$W pump power. By using a pump laser whose pulse width impedance matches with the cavity, those photons are further created in single longitudinal modes with purity reaching 99\%, without relying on later-on filtering. With a dual-channel photon-number resolving detection system, we obtain directly the joint detection probabilities of multi-photon states up to three photons, with high coincidence to accidental contrast for each. Used as a single photon source, it gives heralded $g_H^{(2)}(0)$ around 0.04 at a single photon rate of 650 kHz on chip. The findings of our research highlight the potential of this nanophotonic platform as a promising platform for generating non-classical, few-photon states with ideal indistinguishability, for fundamental quantum optics studies and information applications.

quant-ph↗

arXiv2023

GraspGPT: Leveraging Semantic Knowledge from a Large Language Model for Task-Oriented Grasping

Task-oriented grasping (TOG) refers to the problem of predicting grasps on an object that enable subsequent manipulation tasks. To model the complex relationships between objects, tasks, and grasps, existing methods incorporate semantic knowledge as priors into TOG pipelines. However, the existing semantic knowledge is typically constructed based on closed-world concept sets, restraining the generalization to novel concepts out of the pre-defined sets. To address this issue, we propose GraspGPT, a large language model (LLM) based TOG framework that leverages the open-end semantic knowledge from an LLM to achieve zero-shot generalization to novel concepts. We conduct experiments on Language Augmented TaskGrasp (LA-TaskGrasp) dataset and demonstrate that GraspGPT outperforms existing TOG methods on different held-out settings when generalizing to novel concepts out of the training set. The effectiveness of GraspGPT is further validated in real-robot experiments. Our code, data, appendix, and video are publicly available at https://sites.google.com/view/graspgpt/.

cs.RO↗

arXiv2023

Efficient Object Rearrangement via Multi-view Fusion

The prospect of assistive robots aiding in object organization has always been compelling. In an image-goal setting, the robot rearranges the current scene to match the single image captured from the goal scene. The key to an image-goal rearrangement system is estimating the desired placement pose of each object based on the single goal image and observations from the current scene. In order to establish sufficient associations for accurate estimation, the system should observe an object from a viewpoint similar to that in the goal image. Existing image-goal rearrangement systems, due to their reliance on a fixed viewpoint for perception, often require redundant manipulations to randomly adjust an object's pose for a better perspective. Addressing this inefficiency, we introduce a novel object rearrangement system that employs multi-view fusion. By observing the current scene from multiple viewpoints before manipulating objects, our approach can estimate a more accurate pose without redundant manipulation times. A standard visual localization pipeline at the object level is developed to capitalize on the advantages of multi-view observations. Simulation results demonstrate that the efficiency of our system outperforms existing single-view systems. The effectiveness of our system is further validated in a physical experiment.

cs.RO↗

arXiv2023

An Isotropic Discretization with Semi-implicit Approach for Phase Field Model of Alloy Solidification

Quantitative phase field models have been extensively used to study the solidification behavior of alloys under different conditions. However, a longstanding challenge of phase field models is the directional bias caused by the discretization-induced lattice effects. In particular, widely used discretization methods may introduce significant spurious anisotropy for simulations of polycrystalline solidification. In this paper, we demonstrate a feasible 2D discretization strategy utilizing a hexagonal mesh to reduce the lattice-induced anisotropy of the phase field model. The leading differential terms of the 2D discretization methods are analyzed by using known methods in Fourier space. Using Taylor expansion of discrete Fourier Transform up to sixth order, we found that the proposed discretization strategy is more accurate and isotropic than other methods, including the isotropic discretization recently proposed by Ji et al.[1]. Additionally, the proposed 2D discretization method can be easily incorporated into a semi-implicit algorithm to solve phase field equations, thereby greatly reducing time step constraints and improving computational efficiency compared to explicit approaches. To prove the accuracy and efficiency of the proposed isotropic discretization with semi-implicit algorithm, 2D simulations of alloy solidification with different discretization schemes were performed and compared. We show that the proposed discretization using a hexagonal mesh can drastically reduce grid-induced anisotropy compared to conventional methods.

cond-mat.mtrl-sci↗

arXiv2023

Vertical Ferroelectricity in Van der Waals Materials: Models and Devices

Ferroelectricity has a wide range of applications in functional electronics and is extremely important for the development of next-generation information storage technology, but it is difficult to achieve due to its special symmetry requirements. In this letter, based on van derWaals stacking, a generic model is proposed for realizing ferroelectric devices, where a freely movable center layer is packaged in two fixed and symmetrically stacked layers. In this model, the ferroelectric phase transition can be realized between the two equivalent and eccentric ground stacking-states with opposite polarizations. By means of first-principles calculations, taking the h-BN/h-BN/h-BN and h-BN/Graphene/h-BN as feasible models, we carefully evaluate the magnitude of ferroelectricity. The corresponding polarizations are estimated as 1.83 and 1.35 pC/m, respectively, which are comparable to the sliding ferroelectricity. Such a new tri-layer model of vertical ferroelectricity can be constructed by arbitrary van derWaals semiconducting materials, and usually holds low switching barrier. Optimized material combinations with remarkable polarization are highly expectable to be discovered from the huge candidate set for future information storage.

cond-mat.mtrl-sci↗

arXiv2023

Task-Oriented Grasp Prediction with Visual-Language Inputs

To perform household tasks, assistive robots receive commands in the form of user language instructions for tool manipulation. The initial stage involves selecting the intended tool (i.e., object grounding) and grasping it in a task-oriented manner (i.e., task grounding). Nevertheless, prior researches on visual-language grasping (VLG) focus on object grounding, while disregarding the fine-grained impact of tasks on object grasping. Task-incompatible grasping of a tool will inevitably limit the success of subsequent manipulation steps. Motivated by this problem, this paper proposes GraspCLIP, which addresses the challenge of task grounding in addition to object grounding to enable task-oriented grasp prediction with visual-language inputs. Evaluation on a custom dataset demonstrates that GraspCLIP achieves superior performance over established baselines with object grounding only. The effectiveness of the proposed method is further validated on an assistive robotic arm platform for grasping previously unseen kitchen tools given the task specification. Our presentation video is available at: https://www.youtube.com/watch?v=e1wfYQPeAXU.

cs.RO↗

arXiv2022

Image reconstruction algorithms in radio interferometry: from handcrafted to learned regularization denoisers

We introduce a new class of iterative image reconstruction algorithms for radio interferometry, at the interface of convex optimization and deep learning, inspired by plug-and-play methods. The approach consists in learning a prior image model by training a deep neural network (DNN) as a denoiser, and substituting it for the handcrafted proximal regularization operator of an optimization algorithm. The proposed AIRI (``AI for Regularization in radio-interferometric Imaging'') framework, for imaging complex intensity structure with diffuse and faint emission from visibility data, inherits the robustness and interpretability of optimization, and the learning power and speed of networks. Our approach relies on three steps. Firstly, we design a low dynamic range training database from optical intensity images. Secondly, we train a DNN denoiser at a noise level inferred from the signal-to-noise ratio of the data. We use training losses enhanced with a nonexpansiveness term ensuring algorithm convergence, and including on-the-fly database dynamic range enhancement via exponentiation. Thirdly, we plug the learned denoiser into the forward-backward optimization algorithm, resulting in a simple iterative structure alternating a denoising step with a gradient-descent data-fidelity step. We have validated AIRI against CLEAN, optimization algorithms of the SARA family, and a DNN trained to reconstruct the image directly from visibility data. Simulation results show that AIRI is competitive in imaging quality with SARA and its unconstrained forward-backward-based version uSARA, while providing significant acceleration. CLEAN remains faster but offers lower quality. The end-to-end DNN offers further acceleration, but with far lower quality than AIRI.

eess.IV↗

arXiv2022

Fast and Sensitive Terahertz Detection in a Current-Driven Epitaxial-Graphene Asymmetric Dual-Grating-Gate FET Structure

We designed and fabricated an epitaxial-graphene-channel field-effect transistor (EG-FET) featured by the asymmetric dual-grating-gate (ADGG) structure working for a current-driven terahertz detector, and experimentally demonstrated a 10-ps order fast response time and a high responsivity of 0.3 mA/W to the 0.95-THz radiation incidence at room temperatures. The ADGG- and the drain-source-bias dependencies of the measured photoresponse showed a clear transition between plasmonic detection under periodic electron density modulation conditions with depleted regions and photothermoelectric detection under highly doped conditions without depleted regions. We identified the photothermoelectric detection that we observed as a new type of unipolar mechanism in which only electrons or holes contribute to rectifying the THz radiation under current-driven conditions. These two detection mechanisms coexist in a certain wide transcendent range of the applied bias voltages. It was also clearly manifested that the temporal photoresponse of the plasmonic and photothermoelectric detection are comparably fast on the order of 10 ps, whereas the maximal photoresponsivity of the photothermoelectric detection is almost twice as high as that of the plasmonic detection under the applied biases conditions. These results suggest that the ADGG-EG-FET THz detector will be promising for use in 6G- and 7G-class high-speed wireless communication systems.

physics.optics↗

arXiv2022

MorphoSim: An efficient and scalable phase-field framework for accurately simulating multicellular morphologies

The phase field model can accurately simulate the evolution of microstructures with complex morphologies, and it has been widely used for cell modeling in the last two decades. However, compared to other cellular models such as the coarse-grained model and the vertex model, its high computational cost caused by three-dimensional spatial discretization hampered its application and scalability, especially for multicellular organisms. Recently, we built a phase field model coupled with in vivo imaging data to accurately reconstruct the embryonic morphogenesis of Caenorhabditis elegans from 1- to 8-cell stages [Kuang et al, PLoS Comput. Biol., 2022]. In this work, we propose an improved phase field model by using the stabilized numerical scheme and modified volume constriction. Then we present a scalable phase-field framework, MorphoSim, which is 100 times more efficient than the previous one, and can simulate over 100 mechanically interacting cells. Finally, we demonstrate how MorphoSim can be successfully applied to reproduce the assembly, self-repairing, and dissociation of a synthetic artificial multicellular system - the synNotch system.

physics.bio-ph↗