Searcharxiv⌕ Search

arXiv subjects

Yan Wang

Publications and source records attributed to Yan Wang.

At least 667 records · Page 37Linked to original sources
arXiv2023

Quantum parametric amplifiation of phonon-mediated magnon-spin interaction

The recently developed hybrid magnonics provides new opportunities for advances in both the study of magnetism and the development of quantum information processing. However, engineering coherent quantum state transfer between magnons and specific information carriers, in particular, mechanical oscillators and solid-state spins, remains challenging due to the intrinsically weak interactions and the frequency mismatch between diffrent components. Here, we show how to strongly couple the magnon modes in a nanomagnet to the quantized mechanical motion (phonons) of a micromechanical cantilever in a hybrid tripartite system. The coherent and enhanced magnon-phonon coupling is engineered by introducing the quantum parametric amplifiation of the mechanical motion. With experimentally feasible parameters, we show that the mechanical parametric drive can be adjusted to drive the system into the strong-coupling regime and even the ultrastrong-coupling regime. Furthermore, we show the coherent state transfer between the nanomagnet and a nitrogen-vacancy center in the dispersive-coupling regime, with the magnon-spin interaction mediated by the virtually-excited squeezed phonons. The amplifid mechanical noise can hardly interrupt the coherent dynamics of the system even for low mechanical quality factors, which removes the requirement of applying additional engineered-reservoir techniques. Our work opens up prospects for developing novel quantum transducers, quantum memories and high-precision measurements.

quant-ph↗
arXiv2023

DiffDP: Radiotherapy Dose Prediction via a Diffusion Model

Currently, deep learning (DL) has achieved the automatic prediction of dose distribution in radiotherapy planning, enhancing its efficiency and quality. However, existing methods suffer from the over-smoothing problem for their commonly used L_1 or L_2 loss with posterior average calculations. To alleviate this limitation, we innovatively introduce a diffusion-based dose prediction (DiffDP) model for predicting the radiotherapy dose distribution of cancer patients. Specifically, the DiffDP model contains a forward process and a reverse process. In the forward process, DiffDP gradually transforms dose distribution maps into Gaussian noise by adding small noise and trains a noise predictor to predict the noise added in each timestep. In the reverse process, it removes the noise from the original Gaussian noise in multiple steps with the well-trained noise predictor and finally outputs the predicted dose distribution map. To ensure the accuracy of the prediction, we further design a structure encoder to extract anatomical information from patient anatomy images and enable the noise predictor to be aware of the dose constraints within several essential organs, i.e., the planning target volume and organs at risk. Extensive experiments on an in-house dataset with 130 rectum cancer patients demonstrate the s

eess.IV↗
arXiv2023

Copy Is All You Need

The dominant text generation models compose the output by sequentially selecting words from a fixed vocabulary. In this paper, we formulate text generation as progressively copying text segments (e.g., words or phrases) from an existing text collection. We compute the contextualized representations of meaningful text segments and index them using efficient vector search toolkits. The task of text generation is then decomposed into a series of copy-and-paste operations: at each time step, we seek suitable text spans from the text collection rather than selecting from a standalone vocabulary. Experiments on the standard language modeling benchmark (WikiText-103) show that our approach achieves better generation quality according to both automatic and human evaluations. Besides, its inference efficiency is comparable to token-level autoregressive models thanks to the reduction of decoding steps. We also show that our approach allows for effective domain adaptation by simply switching to domain-specific text collection without extra training. Finally, we observe that our approach attains additional performance gains by simply scaling up to larger text collections, again without further training.\footnote{Our source codes are publicly available at \url{https://github.com/gmftbyGMFTBY/Copyisallyouneed}.}

cs.CL↗
arXiv2023

Graphs with girth 9 and without longer odd holes are 3-colorable

For a number $l\geq 2$, let ${\cal{G}}_l$ denote the family of graphs which have girth $2l+1$ and have no odd hole with length greater than $2l+1$. Wu, Xu and Xu conjectured that every graph in $\bigcup_{l\geq 2} {\cal{G}}_{l}$ is $3$-colorable. Chudnovsky et al., Wu et al., and Chen showed that every graph in ${\cal{G}}_2$, ${\cal{G}}_3$ and $\bigcup_{l\geq 5} {\cal{G}}_{l}$ is $3$-colorable respectively. In this paper, we prove that every graph in ${\cal{G}}_4$ is $3$-colorable. This confirms Wu, Xu and Xu's conjecture.

math.CO↗
arXiv2023

SysNoise: Exploring and Benchmarking Training-Deployment System Inconsistency

Extensive studies have shown that deep learning models are vulnerable to adversarial and natural noises, yet little is known about model robustness on noises caused by different system implementations. In this paper, we for the first time introduce SysNoise, a frequently occurred but often overlooked noise in the deep learning training-deployment cycle. In particular, SysNoise happens when the source training system switches to a disparate target system in deployments, where various tiny system mismatch adds up to a non-negligible difference. We first identify and classify SysNoise into three categories based on the inference stage; we then build a holistic benchmark to quantitatively measure the impact of SysNoise on 20+ models, comprehending image classification, object detection, instance segmentation and natural language processing tasks. Our extensive experiments revealed that SysNoise could bring certain impacts on model robustness across different tasks and common mitigations like data augmentation and adversarial training show limited effects on it. Together, our findings open a new research topic and we hope this work will raise research attention to deep learning deployment systems accounting for model performance. We have open-sourced the benchmark and framework at https://modeltc.github.io/systemnoise_web.

cs.LG↗
arXiv2023

Improving the Transferability of Time Series Forecasting with Decomposition Adaptation

Due to effective pattern mining and feature representation, neural forecasting models based on deep learning have achieved great progress. The premise of effective learning is to collect sufficient data. However, in time series forecasting, it is difficult to obtain enough data, which limits the performance of neural forecasting models. To alleviate the data scarcity limitation, we design Sequence Decomposition Adaptation Network (SeDAN) which is a novel transfer architecture to improve forecasting performance on the target domain by aligning transferable knowledge from cross-domain datasets. Rethinking the transferability of features in time series data, we propose Implicit Contrastive Decomposition to decompose the original features into components including seasonal and trend features, which are easier to transfer. Then we design the corresponding adaptation methods for decomposed features in different domains. Specifically, for seasonal features, we perform joint distribution adaptation and for trend features, we design an Optimal Local Adaptation. We conduct extensive experiments on five benchmark datasets for multivariate time series forecasting. The results demonstrate the effectiveness of our SeDAN. It can provide more efficient and stable knowledge transfer.

cs.LG↗
arXiv2023

Mechanical investigations of composite cathode degradation in all-solid-state-batteries

Despite ongoing efforts aimed at increasing energy density in all-solid-state-batteries, the optimal composite cathode morphology, which requires minimal volume change, small void development, and good interfacial contact, remains a significant concern within the community. In this work, we focus on the theoretical investigation of the above-mentioned mechanical defects in the composite cathode during electrochemical cycling. It is demonstrated that these mechanical defects are highly dependent on the SE material properties, the external stack pressure and the cathode active material (CAM) loading. The following conclusions are highlighted in this study: (1) Higher CAM loading (>50 vol. %) causes an increase in mechanical defects, including large cathode volume change (>5%), contact loss (50%) and porosity (>1%). (2) High external stack pressure up to 7MPa reduces mechanical defects while preventing internal fracture in the cathode. (3) Soft SE materials with small Youngs modulus (<10GPa) and low hardness (<2GPa) can significantly minimize these mechanical defects during cycling. (4) A design strategy is proposed for high CAM loading with minimal mechanical defects when different SE materials are utilized in the composite cathode, including oxide-type SE, sulfide-type SE, and halide-type SE. The research provides specific guidelines to optimize the composite cathode in terms of mechanical properties. These guidelines broaden the design approach towards improving the performance of SSB, by highlighting the importance of considering the mechanical properties of battery materials.

cond-mat.mtrl-sci↗
arXiv2023

A Unified Framework for Online Data-Driven Predictive Control with Robust Safety Guarantees

Despite great successes, model predictive control (MPC) relies on an accurate dynamical model and requires high onboard computational power, impeding its wider adoption in engineering systems, especially for nonlinear real-time systems with limited computation power. These shortcomings of MPC motivate this work to make such a control framework more practically viable for real-world applications. Specifically, to remove the required accurate dynamical model and reduce the computational cost for nonlinear MPC (NMPC), this paper develops a unified online data-driven predictive control pipeline to efficiently control a system with guaranteed safety without incurring large computational complexity. The new aspect of this idea is learning not only the real system but also the control policy, which results in a reasonable computational cost for the data-driven predictive controllers. More specifically, we first develop a spatial temporal filter (STF)-based concurrent learning scheme to systematically identify system dynamics for general nonlinear systems. We then develop a robust control barrier function (RCBF) for safety guarantees in the presence of model uncertainties and learn the RCBF-based NMPC policy. Furthermore, to mitigate the performance degradation due to the existing model uncertainties, we propose an online policy correction scheme through perturbation analysis and design of an ancillary feedback controller. Finally, extensive simulations on two applications, cart-inverted pendulum and automotive powertrain control, are performed to demonstrate the efficacy of the proposed framework, which shows comparable performance with much lower computational cost in comparison with several benchmark algorithms.

eess.SY↗
arXiv2023

Real-Time Reinforcement Learning for Vision-Based Robotics Utilizing Local and Remote Computers

Real-time learning is crucial for robotic agents adapting to ever-changing, non-stationary environments. A common setup for a robotic agent is to have two different computers simultaneously: a resource-limited local computer tethered to the robot and a powerful remote computer connected wirelessly. Given such a setup, it is unclear to what extent the performance of a learning system can be affected by resource limitations and how to efficiently use the wirelessly connected powerful computer to compensate for any performance loss. In this paper, we implement a real-time learning system called the Remote-Local Distributed (ReLoD) system to distribute computations of two deep reinforcement learning (RL) algorithms, Soft Actor-Critic (SAC) and Proximal Policy Optimization (PPO), between a local and a remote computer. The performance of the system is evaluated on two vision-based control tasks developed using a robotic arm and a mobile robot. Our results show that SAC's performance degrades heavily on a resource-limited local computer. Strikingly, when all computations of the learning system are deployed on a remote workstation, SAC fails to compensate for the performance loss, indicating that, without careful consideration, using a powerful remote computer may not result in performance improvement. However, a carefully chosen distribution of computations of SAC consistently and substantially improves its performance on both tasks. On the other hand, the performance of PPO remains largely unaffected by the distribution of computations. In addition, when all computations happen solely on a powerful tethered computer, the performance of our system remains on par with an existing system that is well-tuned for using a single machine. ReLoD is the only publicly available system for real-time RL that applies to multiple robots for vision-based tasks.

cs.RO↗
arXiv2023

The space cold atom interferometer for testing the equivalence principle in the China Space Station

The precision of the weak equivalence principle (WEP) test using atom interferometers (AIs) is expected to be extremely high in microgravity environment. The microgravity scientific laboratory cabinet (MSLC) in the China Space Station (CSS) can provide a higher-level microgravity than the CSS itself, which provides a good experimental environment for scientific experiments that require high microgravity. We designed and realized a payload of a dual-species cold rubidium atom interferometer. The payload is highly integrated and has a size of 460 mm * 330 mm * 260 mm. It will be installed in the MSLC to carry out high-precision WEP test experiment. In this article, we introduce the constraints and guidelines of the payload design, the compositions and functions of the scientific payload, the expected test precision in space, and some results of the ground test experiments

physics.atom-ph↗
arXiv2023

Gravitational waveforms from the inspiral of compact binaries in the Brans-Dicke theory in an expanding Universe

In modified gravity theories, such as the Brans-Dicke theory, the background evolution of the Universe and the perturbation around it are different from that in general relativity. Therefore, the gravitational waveforms used to study standard sirens in these theories should be modified. The modifications of the waveforms can be classified into two categories: wave generation effects and wave propagation effects. Hitherto, the waveforms used to study standard sirens in the modified gravity theories incorporate only the wave propagation effects and ignore the wave generation effects; while the waveforms focusing on the wave generation effects, such as the post-Newtonian waveforms, do not incorporate the wave propagation effects and cannot be directly applied to the sources with non-negligible redshifts in the study of standard sirens. In this work, we construct the consistent waveforms for standard sirens in the Brans-Dicke theory. The wave generation effects include the emission of the scalar breathing polarization $h_b$ and the corrections to the tensor polarizations $h_+$ and $h_\times$; the wave propagation effect is the modification of the luminosity distance for the gravitational waveforms. Using the consistent waveforms, we analyze the parameter estimation biases due to the ignorance of the wave generation effects. Considering the observations by the Einstein Telescope, we find that the ratio of the theoretical bias to the statistical error of the redshifted chirp mass is two orders of magnitude larger than that of the source distance. For black hole-neutron star binary systems like GW191219, the theoretical bias of the redshifted chirp mass can be several times larger than the statistical error.

gr-qc↗
arXiv2023

MedNgage: A Dataset for Understanding Engagement in Patient-Nurse Conversations

Patients who effectively manage their symptoms often demonstrate higher levels of engagement in conversations and interventions with healthcare practitioners. This engagement is multifaceted, encompassing cognitive and socio-affective dimensions. Consequently, it is crucial for AI systems to understand the engagement in natural conversations between patients and practitioners to better contribute toward patient care. In this paper, we present a novel dataset (MedNgage), which consists of patient-nurse conversations about cancer symptom management. We manually annotate the dataset with a novel framework of categories of patient engagement from two different angles, namely: i) socio-affective (3.1K spans), and ii) cognitive use of language (1.8K spans). Through statistical analysis of the data that is annotated using our framework, we show a positive correlation between patient symptom management outcomes and their engagement in conversations. Additionally, we demonstrate that pre-trained transformer models fine-tuned on our dataset can reliably predict engagement classes in patient-nurse conversations. Lastly, we use LIME (Ribeiro et al., 2016) to analyze the underlying challenges of the tasks that state-of-the-art transformer models encounter. The de-identified data is available for research purposes upon request.

cs.CL↗
arXiv2023

Extended Neighboring Extremal Optimal Control with State and Preview Perturbations

Optimal control schemes have achieved remarkable performance in numerous engineering applications. However, they typically require high computational cost, which has limited their use in real-world engineering systems with fast dynamics and/or limited computation power. To address this challenge, Neighboring Extremal (NE) has been developed as an efficient optimal adaption strategy to adapt a pre-computed nominal control solution to perturbations from the nominal trajectory. The resulting control law is a time-varying feedback gain that can be pre-computed along with the original optimal control problem, and it takes negligible online computation. However, existing NE frameworks only deal with state perturbations while in modern applications, optimal controllers (e.g., predictive controllers) frequently incorporate preview information. Therefore, a new NE framework is needed to adapt to such preview perturbations. In this work, an extended NE (ENE) framework is developed to systematically adapt the nominal control to both state and preview perturbations. We show that the derived ENE law is two time-varying feedback gains on the state perturbation and the preview perturbation. We also develop schemes to handle nominal non-optimal solutions and large perturbations to retain optimal performance and constraint satisfaction. Case study on nonlinear model predictive control is presented due to its popularity but it can be easily extended to other optimal control schemes. Promising simulation results on the cart inverted pendulum problem demonstrate the efficacy of the ENE algorithm.

eess.SY↗
arXiv2023

Asymptotic Performance Analysis of Large-Scale Active IRS-Aided Wireless Network

In this paper, the dominant factor affecting the performance of active intelligent reflecting surface (IRS) aided wireless communication networks in Rayleigh fading channel, namely the average signal-to-noise ratio (SNR) $γ_0$ at IRS, is studied. Making use of the weak law of large numbers, its simple asymptotic expression is derived as the number $N$ of IRS elements goes to medium-scale and large-scale. When $N$ tends to large-scale, the asymptotic received SNR at user is proved to be a linear increasing function of a product of $γ_0$ and $N$. Subsequently, when the BS transmit power is fixed, there exists an optimal limited reflective power at IRS. At this point, more IRS reflect power will degrade the SNR performance. Additionally, under the total power sum constraint of the BS transmit power and the power reflected by the IRS, an optimal power allocation (PA) strategy is derived and shown to achieve 0.83 bit rate gain over equal PA. Finally, an IRS with finite phase shifters being taken into account, generates phase quantization errors, and further leads to a degradation of receive performance. The corresponding closed-form performance loss expressions for user's asymptotic SNR, achievable rate (AR), and bit error rate (BER) are derived for active IRS. Numerical simulation results show that a 3-bit discrete phase shifter is required to achieve a trivial performance loss for a large-scale active IRS.

eess.SP↗
arXiv2023

IB-UQ: Information bottleneck based uncertainty quantification for neural function regression and neural operator learning

We propose a novel framework for uncertainty quantification via information bottleneck (IB-UQ) for scientific machine learning tasks, including deep neural network (DNN) regression and neural operator learning (DeepONet). Specifically, we incorporate the bottleneck by a confidence-aware encoder, which encodes inputs into latent representations according to the confidence of the input data belonging to the region where training data is located, and utilize a Gaussian decoder to predict means and variances of outputs conditional on representation variables. Furthermore, we propose a data augmentation based information bottleneck objective which can enhance the quantification quality of the extrapolation uncertainty, and the encoder and decoder can be both trained by minimizing a tractable variational bound of the objective. In comparison to uncertainty quantification (UQ) methods for scientific learning tasks that rely on Bayesian neural networks with Hamiltonian Monte Carlo posterior estimators, the model we propose is computationally efficient, particularly when dealing with large-scale data sets. The effectiveness of the IB-UQ model has been demonstrated through several representative examples, such as regression for discontinuous functions, real-world data set regression, learning nonlinear operators for partial differential equations, and a large-scale climate model. The experimental results indicate that the IB-UQ model can handle noisy data, generate robust predictions, and provide confident uncertainty evaluation for out-of-distribution data.

math.NA↗
arXiv2023

Joint Uplink and Downlink Resource Allocation Towards Energy-efficient Transmission for URLLC

Ultra-reliable and low-latency communications (URLLC) is firstly proposed in 5G networks, and expected to support applications with the most stringent quality-of-service (QoS). However, since the wireless channels vary dynamically, the transmit power for ensuring the QoS requirements of URLLC may be very high, which conflicts with the power limitation of a real system. To fulfill the successful URLLC transmission with finite transmit power, we propose an energy-efficient packet delivery mechanism incorparated with frequency-hopping and proactive dropping in this paper. To reduce uplink outage probability, frequency-hopping provides more chances for transmission so that the failure hardly occurs. To avoid downlink outage from queue clearing, proactive dropping controls overall reliability by introducing an extra error component. With the proposed packet delivery mechanism, we jointly optimize bandwidth allocation and power control of uplink and downlink, antenna configuration, and subchannel assignment to minimize the average total power under the constraint of URLLC transmission requirements. Via theoretical analysis (e.g., the convexity with respect to bandwidth, the independence of bandwidth allocation, the convexity of antenna configuration with inactive constraints), the simplication of finding the global optimal solution for resource allocation is addressed. A three-step method is then proposed to find the optimal solution for resource allocation. Simulation results validate the analysis and show the performance gain by optimizing resource allocation with the proposed packet delivery mechanism.

eess.SP↗
arXiv2023

PandaGPT: One Model To Instruction-Follow Them All

We present PandaGPT, an approach to emPower large lANguage moDels with visual and Auditory instruction-following capabilities. Our pilot experiments show that PandaGPT can perform complex tasks such as detailed image description generation, writing stories inspired by videos, and answering questions about audios. More interestingly, PandaGPT can take multimodal inputs simultaneously and compose their semantics naturally. For example, PandaGPT can connect how objects look in an image/video and how they sound in an audio. To do so, PandaGPT combines the multimodal encoders from ImageBind and the large language models from Vicuna. Notably, only aligned image-text pairs are required for the training of PandaGPT. Thanks to the strong capability of ImageBind in embedding data from different modalities into the same space, PandaGPT displays emergent, i.e. zero-shot, cross-modal behaviors for data other than image and text (e.g., video, audio, depth, thermal, and IMU). We hope that PandaGPT serves as an initial step toward building AGI that can perceive and understand inputs in different modalities holistically, as we humans do. Our project page is at https://panda-gpt.github.io/.

cs.CL↗
arXiv2023

Sequential Latin Hypercube Design for Two-layer Computer Simulators

The two-layer computer simulators are commonly used to mimic multi-physics phenomena or systems. Usually, the outputs of the first-layer simulator (also called the inner simulator) are partial inputs of the second-layer simulator (also called the outer simulator). How to design experiments by considering the space-filling properties of inner and outer simulators simultaneously is a significant challenge that has received scant attention in the literature. To address this problem, we propose a new sequential optimal Latin hypercube design (LHD) by using the maximin integrating mixed distance criterion. A corresponding sequential algorithm for efficiently generating such designs is also developed. Numerical simulation results show that the new method can effectively improve the space-filling property of the outer computer inputs. The case study about composite structures assembly simulation demonstrates that the proposed method can outperform the benchmark methods.

stat.ME↗