arXiv subjects

Shuai Yuan

Publications and source records attributed to Shuai Yuan.

At least 91 records · Page 5Linked to original sources

arXiv2023

SemARFlow: Injecting Semantics into Unsupervised Optical Flow Estimation for Autonomous Driving

Unsupervised optical flow estimation is especially hard near occlusions and motion boundaries and in low-texture regions. We show that additional information such as semantics and domain knowledge can help better constrain this problem. We introduce SemARFlow, an unsupervised optical flow network designed for autonomous driving data that takes estimated semantic segmentation masks as additional inputs. This additional information is injected into the encoder and into a learned upsampler that refines the flow output. In addition, a simple yet effective semantic augmentation module provides self-supervision when learning flow and its boundaries for vehicles, poles, and sky. Together, these injections of semantic information improve the KITTI-2015 optical flow test error rate from 11.80% to 8.38%. We also show visible improvements around object boundaries as well as a greater ability to generalize across datasets. Code is available at https://github.com/duke-vision/semantic-unsup-flow-release.

cs.CV↗

arXiv2023

Mitigating Test-Time Bias for Fair Image Retrieval

We address the challenge of generating fair and unbiased image retrieval results given neutral textual queries (with no explicit gender or race connotations), while maintaining the utility (performance) of the underlying vision-language (VL) model. Previous methods aim to disentangle learned representations of images and text queries from gender and racial characteristics. However, we show these are inadequate at alleviating bias for the desired equal representation result, as there usually exists test-time bias in the target retrieval set. So motivated, we introduce a straightforward technique, Post-hoc Bias Mitigation (PBM), that post-processes the outputs from the pre-trained vision-language model. We evaluate our algorithm on real-world image search datasets, Occupation 1 and 2, as well as two large-scale image-text datasets, MS-COCO and Flickr30k. Our approach achieves the lowest bias, compared with various existing bias-mitigation methods, in text-based image retrieval result while maintaining satisfactory retrieval performance. The source code is publicly available at \url{https://anonymous.4open.science/r/Fair_Text_based_Image_Retrieval-D8B2}.

cs.CV↗

arXiv2023

Hexagonal close-packed polar-skyrmion lattice in ultrathin ferroelectric PbTiO3 films

Polar skyrmions are topologically stable, swirling polarization textures with particle-like characteristics, which hold promise for next-generation, nanoscale logic and memory. While understanding of how to create ordered polar skyrmion lattice structures and how such structure respond to applied electric fields, temperature, and film thickness remains elusive. Here, using phase-field simulations, the evolution of polar topology and the emergence of a phase transition to a hexagonal close-packed skyrmion lattice is explored through the construction of a temperature-electric field phase diagram for ultrathin ferroelectric PbTiO3 films. The hexagonal-lattice skyrmion crystal can be stabilized under application of an external, out-of-plane electric field which carefully adjusts the delicate interplay of elastic, electrostatic, and gradient energies. In addition, the lattice constants of the polar skyrmion crystals are found to increase with film thickness, consistent with expectation from Kittel law. Our studies pave the way for the development of novel ordered condensed matter phases assembled from topological polar textures and related emergent properties in nanoscale ferroelectrics.

cond-mat.mtrl-sci↗

arXiv2023

ERRA: An Embodied Representation and Reasoning Architecture for Long-horizon Language-conditioned Manipulation Tasks

This letter introduces ERRA, an embodied learning architecture that enables robots to jointly obtain three fundamental capabilities (reasoning, planning, and interaction) for solving long-horizon language-conditioned manipulation tasks. ERRA is based on tightly-coupled probabilistic inferences at two granularity levels. Coarse-resolution inference is formulated as sequence generation through a large language model, which infers action language from natural language instruction and environment state. The robot then zooms to the fine-resolution inference part to perform the concrete action corresponding to the action language. Fine-resolution inference is constructed as a Markov decision process, which takes action language and environmental sensing as observations and outputs the action. The results of action execution in environments provide feedback for subsequent coarse-resolution reasoning. Such coarse-to-fine inference allows the robot to decompose and achieve long-horizon tasks interactively. In extensive experiments, we show that ERRA can complete various long-horizon manipulation tasks specified by abstract language instructions. We also demonstrate successful generalization to the novel but similar natural language instructions.

cs.RO↗

arXiv2023

Anomalous quantized plateaus in two-dimensional electron gas with gate confinement

Quantum information can be coded by the topologically protected edges of fractional quantum Hall (FQH) states. Investigation on FQH edges in the hope of searching and utilizing non-Abelian statistics has been a focused challenge for years. Manipulating the edges, e.g. to bring edges close to each other or to separate edges spatially, is a common and essential step for such studies. The FQH edge structures in a confined region are typically presupposed to be the same as that in the open region in analysis of experimental results, but whether they remain unchanged with extra confinement is obscure. In this work, we present a series of unexpected plateaus in a confined single-layer two-dimensional electron gas (2DEG), which are quantized at anomalous fractions such as 9/4, 17/11, 16/13 and the reported 3/2. We explain all the plateaus by assuming surprisingly larger filling factors in the confined region. Our findings enrich the understanding of edge states in the confined region and in the applications of gate manipulation, which is crucial for the experiments with quantum point contact and interferometer.

cond-mat.mes-hall↗

arXiv2022

Cross-Attention Transformer for Video Interpolation

We propose TAIN (Transformers and Attention for video INterpolation), a residual neural network for video interpolation, which aims to interpolate an intermediate frame given two consecutive image frames around it. We first present a novel vision transformer module, named Cross Similarity (CS), to globally aggregate input image features with similar appearance as those of the predicted interpolated frame. These CS features are then used to refine the interpolated prediction. To account for occlusions in the CS features, we propose an Image Attention (IA) module to allow the network to focus on CS features from one frame over those of the other. TAIN outperforms existing methods that do not require flow estimation and performs comparably to flow-based methods while being computationally efficient in terms of inference time on Vimeo90k, UCF101, and SNU-FILM benchmarks.

cs.CV↗

arXiv2022

Optical Flow Training under Limited Label Budget via Active Learning

Supervised training of optical flow predictors generally yields better accuracy than unsupervised training. However, the improved performance comes at an often high annotation cost. Semi-supervised training trades off accuracy against annotation cost. We use a simple yet effective semi-supervised training method to show that even a small fraction of labels can improve flow accuracy by a significant margin over unsupervised training. In addition, we propose active learning methods based on simple heuristics to further reduce the number of labels required to achieve the same target accuracy. Our experiments on both synthetic and real optical flow datasets show that our semi-supervised networks generally need around 50% of the labels to achieve close to full-label accuracy, and only around 20% with active learning on Sintel. We also analyze and show insights on the factors that may influence active learning performance. Code is available at https://github.com/duke-vision/optical-flow-active-learning-release.

cs.CV↗

arXiv2022

Unsupervised Flow Refinement near Motion Boundaries

Unsupervised optical flow estimators based on deep learning have attracted increasing attention due to the cost and difficulty of annotating for ground truth. Although performance measured by average End-Point Error (EPE) has improved over the years, flow estimates are still poorer along motion boundaries (MBs), where the flow is not smooth, as is typically assumed, and where features computed by neural networks are contaminated by multiple motions. To improve flow in the unsupervised settings, we design a framework that detects MBs by analyzing visual changes along boundary candidates and replaces motions close to detections with motions farther away. Our proposed algorithm detects boundaries more accurately than a baseline method with the same inputs and can improve estimates from any flow predictor without additional training.

cs.CV↗

arXiv2022

Thin-Film Lithium Niobate based Dual-Polarization IQ modulator for Single-Carrier 1.6 Tb/s Transmission

We successfully demonstrate a monolithic integrated dual-polarization (DP) IQ modulator based on thin-film lithium niobate (TFLN) platform with a silicon substrate, which consists of IQ modulators, spot-size converters (SSCs) and a polarization rotator combiner (PRC). After coupled with polarization maintaining fibers, the measured insertion loss of the modulator is 12 dB. In addition, we experimentally achieve a single-carrier 1.6 Tb/s net bitrate transmission.

physics.app-ph↗

arXiv2022

Adversarial Jamming for a More Effective Constellation Attack

The common jamming mode in wireless communication is band barrage jamming, which is controllable and difficult to resist. Although this method is simple to implement, it is obviously not the best jamming waveform. Therefore, based on the idea of adversarial examples, we propose the adversarial jamming waveform, which can independently optimize and find the best jamming waveform. We attack QAM with adversarial jamming and find that the optimal jamming waveform is equivalent to the amplitude and phase between the nearest constellation points. Furthermore, by verifying the jamming performance on a hardware platform, it is shown that our method significantly improves the bit error rate compared to other methods.

cs.CR↗

arXiv2021

Guarantees for Tuning the Step Size using a Learning-to-Learn Approach

Choosing the right parameters for optimization algorithms is often the key to their success in practice. Solving this problem using a learning-to-learn approach -- using meta-gradient descent on a meta-objective based on the trajectory that the optimizer generates -- was recently shown to be effective. However, the meta-optimization problem is difficult. In particular, the meta-gradient can often explode/vanish, and the learned optimizer may not have good generalization performance if the meta-objective is not chosen carefully. In this paper we give meta-optimization guarantees for the learning-to-learn approach on a simple problem of tuning the step size for quadratic loss. Our results show that the naïve objective suffers from meta-gradient explosion/vanishing problem. Although there is a way to design the meta-objective so that the meta-gradient remains polynomially bounded, computing the meta-gradient directly using backpropagation leads to numerical issues. We also characterize when it is necessary to compute the meta-objective on a separate validation set to ensure the generalization performance of the learned optimizer. Finally, we verify our results empirically and show that a similar phenomenon appears even for more complicated learned optimizers parametrized by neural networks.

stat.ML↗

arXiv2021

RadioNet: Transformer based Radio Map Prediction Model For Dense Urban Environments

Radio Map Prediction (RMP), aiming at estimating coverage of radio wave, has been widely recognized as an enabling technology for improving radio spectrum efficiency. However, fast and reliable radio map prediction can be very challenging due to the complicated interaction between radio waves and the environment. In this paper, a novel Transformer based deep learning model termed as RadioNet is proposed for radio map prediction in urban scenarios. In addition, a novel Grid Embedding technique is proposed to substitute the original Position Embedding in Transformer to better anchor the relative position of the radiation source, destination and environment. The effectiveness of proposed method is verified on an urban radio wave propagation dataset. Compared with the SOTA model on RMP task, RadioNet reduces the validation loss by 27.3\%, improves the prediction reliability from 90.9\% to 98.9\%. The prediction speed is increased by 4 orders of magnitude, when compared with ray-tracing based method. We believe that the proposed method will be beneficial to high-efficiency wireless communication, real-time radio visualization, and even high-speed image rendering.

eess.SP↗

arXiv2021

Mean Field MARL Based Bandwidth Negotiation Method for Massive Devices Spectrum Sharing

In this paper, a novel bandwidth negotiation mechanism is proposed for massive devices wireless spectrum sharing, in which individual device locally negotiates bandwidth usage with neighbor devices and globally optimal spectrum utilization is achieved through distributed decision-making. Since only sparse feedback is needed, the proposed mechanism can greatly reduce the signaling overhead. In order to solve the distributed optimization problem when massive devices coexist, mean field multi-agent reinforcement learning (MF-MARL) based bandwidth decision algorithm is proposed, which allow device make globally optimal decision leveraging only neighborhood observation. In simulation, distributed bandwidth negotiation between 1000 devices is demonstrated and the spectrum utilization rate is above 95%. The proposed method is beneficial to reduce spectrum conflicts, increase spectrum utilization for massive devices spectrum sharing.

eess.SP↗

arXiv2021

Noise Attention based Spectrum Anomaly Detection Method for Unauthorized Bands

Spectrum anomaly detection is of great importance in wireless communication to secure safety and improve spectrum efficiency. However, spectrum anomaly detection faces many difficulties, especially in unauthorized frequency bands. For example, the composition of unauthorized frequency bands is very complex and the abnormal usage patterns are unknown in prior. In this paper, a noise attention method is proposed for unsupervised spectrum anomaly detection in unauthorized bands. First of all, we theoretically prove that the anomalies in unauthorized bands will raise the noise floor of spectrogram after VAE reconstruction. Then, we introduce a novel anomaly metric named as noise attention score to more effectively capture spectrum anomaly. The effectiveness of the proposed method is experimentally verified in 2.4 GHz ISM band. Leveraging the noise attention score, the AUC metric of anomaly detection is increased by 0.193. The proposed method is beneficial to reliably detecting abnormal spectrum while keeping low false alarm rate.

eess.SP↗

arXiv2021

Extremum seeking control of a class of constrained nonlinear systems

This paper studies the extremum seeking control (ESC) problem for a class of constrained nonlinear systems. Specifically, we focus on a family of constraints allowing to reformulate the original nonlinear system in the so-called input-output normal form. To steer the system to optimize a performance function without knowing its explicit form, we propose a novel numerical optimization-based extremum seeking control (NOESC) design consisting of a constrained numerical optimization method and an inversion based feedforward controller. In particular, a projected gradient descent algorithm is exploited to produce the state sequence to optimize the performance function, whereas a suitable boundary value problem accommodates the finite-time state transition between each two consecutive points of the state sequence. Compared to available NOESC methods, the proposed approach i) can explicitly deal with output constraints; ii) the performance function can consider a direct dependence on the states of the internal dynamics; iii) the internal dynamics do not have to be necessarily stable. The effectiveness of the proposed ESC scheme is shown through extensive numerical simulations.

math.OC↗

arXiv2020

The noncommutativity of the static and homogeneous limit of the axial chemical potential in chiral magnetic effect

We study the noncommutativity of different orders of zero energy-momentum limit pertaining to the axial chemical potential in the chiral magnetic effect. While this noncommutativity issue originates from the pinching singularity at one-loop order, it cannot be removed by introducing a damping term to the fermion propagators. The physical reason is that modifying the propagator alone would violate the axial-vector Ward identity and as a result a modification of the longitudinal component of the axial-vector vertex is required, which contributes to CME. The pinching singularity with free fermion propagators was then taken over by the singularity stemming from the dressed axial-vector vertex. We show this mechanism by a concrete example. Moreover, we proved in general the vanishing CME in the limit order that the static limit was taken prior to the homogeneous limit in the light of Coleman-Hill theorem for a static external magnetic field. For the opposite limit that the homogeneous limit is taken first, we show that the nonvanishing CME was a consequence of the nonrenormalizability of chiral anomaly for an arbitrary external magnetic field.

hep-ph↗

arXiv2019

Spatially Constrained Spectral Clustering Algorithms for Region Delineation

Regionalization is the task of dividing up a landscape into homogeneous patches with similar properties. Although this task has a wide range of applications, it has two notable challenges. First, it is assumed that the resulting regions are both homogeneous and spatially contiguous. Second, it is well-recognized that landscapes are hierarchical such that fine-scale regions are nested wholly within broader-scale regions. To address these two challenges, first, we develop a spatially constrained spectral clustering framework for region delineation that incorporates the tradeoff between region homogeneity and spatial contiguity. The framework uses a flexible, truncated exponential kernel to represent the spatial contiguity constraints, which is integrated with the landscape feature similarity matrix for region delineation. To address the second challenge, we extend the framework to create fine-scale regions that are nested within broader-scaled regions using a greedy, recursive bisection approach. We present a case study of a terrestrial ecology data set in the United States that compares the proposed framework with several baseline methods for regionalization. Experimental results suggest that the proposed framework for regionalization outperforms the baseline methods, especially in terms of balancing region contiguity and homogeneity, as well as creating regions of more similar size, which is often a desired trait of regions.

cs.LG↗

arXiv2019

High efficiency resonant-metasurface germanium photodetector with ultra-thin intrinsic layer

Photodetectors at telecommunications-band with high efficiency and high speed are becoming increasingly important as the booming of big data, 5G, internet of things, cloud computing, artificial intelligence and relevent applications. Silicon-based Germanium photodetectors exhibit great potential in reducing the cost and power dissipation, due to its compatibility of monolithic integration with signal-processing electronics. We report the first demonstration of normal incident resonant-metasurface germanium photodetector, to address the trade-off between quantum efficiency, bandwidth and wavelength coverage for free-space detectors. With an ultra-thin intrinsic layer thickness of 350 nm, a high external quantum efficiency of more than 60% and clearly open eyes at the speed of 20 Gbps are achieved, for a 30 μm-diameter device. The photodetector employs multiple trapped-mode resonances to enhance the localized electromagnetic field, which not only enhances the external quantum efficiency by more than 300% at 1550 nm, but also extends to the whole C band. Further simulation and measurement with small incident spot size show the feasibility to achieve >50 GHz 3 dB bandwidth by simply reducing the mesa size, with minor sacrifice of the enhanced absorption. This work paves the way for the future development of low-cost, high efficiency normal incident germanium photodetectors operating at data rates of 50 Gbps or higher.

physics.ins-det↗