Searcharxiv⌕ Search

arXiv subjects

Peng Chen

Publications and source records attributed to Peng Chen.

At least 235 records · Page 13Linked to original sources
arXiv2023

Enabling Secure Wireless Communications via Movable Antennas

A pioneering secure transmission scheme is proposed, which harnesses movable antennas (MAs) to optimize antenna positions for augmenting the physical layer security. Particularly, an MA-enabled secure wireless system is considered, where a multi-antenna transmitter communicates with a single-antenna receiver in the presence of an eavesdropper. The beamformer and antenna positions at the transmitter are jointly optimized under two criteria: power consumption minimization and secrecy rate maximization. For each scenario, a novel suboptimal algorithm was proposed to tackle the resulting nonconvex optimization problem, capitalizing on the approaches of alternating optimization and gradient descent. Numerical results demonstrate that the proposed MA systems significantly improve physical layer security compared to various benchmark schemes relying on conventional fixed-position antennas (FPAs).

eess.SP↗
arXiv2023

Taiyi: A Bilingual Fine-Tuned Large Language Model for Diverse Biomedical Tasks

Objective: Most existing fine-tuned biomedical large language models (LLMs) focus on enhancing performance in monolingual biomedical question answering and conversation tasks. To investigate the effectiveness of the fine-tuned LLMs on diverse biomedical NLP tasks in different languages, We present Taiyi, a bilingual fine-tuned LLM for diverse biomedical tasks. Materials and Methods: We first curated a comprehensive collection of 140 existing biomedical text mining datasets (102 English and 38 Chinese datasets) across over 10 task types. Subsequently, a two-stage strategy is proposed for supervised fine-tuning to optimize the model performance across varied tasks. Results: Experimental results on 13 test sets covering named entity recognition, relation extraction, text classification, question answering tasks demonstrate that Taiyi achieves superior performance compared to general LLMs. The case study involving additional biomedical NLP tasks further shows Taiyi's considerable potential for bilingual biomedical multi-tasking. Conclusion: Leveraging rich high-quality biomedical corpora and developing effective fine-tuning strategies can significantly improve the performance of LLMs within the biomedical domain. Taiyi shows the bilingual multi-tasking capability through supervised fine-tuning. However, those tasks such as information extraction that are not generation tasks in nature remain challenging for LLM-based generative approaches, and they still underperform the conventional discriminative approaches of smaller language models.

cs.CL↗
arXiv2023

DiffusionTalker: Personalization and Acceleration for Speech-Driven 3D Face Diffuser

Speech-driven 3D facial animation has been an attractive task in both academia and industry. Traditional methods mostly focus on learning a deterministic mapping from speech to animation. Recent approaches start to consider the non-deterministic fact of speech-driven 3D face animation and employ the diffusion model for the task. However, personalizing facial animation and accelerating animation generation are still two major limitations of existing diffusion-based methods. To address the above limitations, we propose DiffusionTalker, a diffusion-based method that utilizes contrastive learning to personalize 3D facial animation and knowledge distillation to accelerate 3D animation generation. Specifically, to enable personalization, we introduce a learnable talking identity to aggregate knowledge in audio sequences. The proposed identity embeddings extract customized facial cues across different people in a contrastive learning manner. During inference, users can obtain personalized facial animation based on input audio, reflecting a specific talking style. With a trained diffusion model with hundreds of steps, we distill it into a lightweight model with 8 steps for acceleration. Extensive experiments are conducted to demonstrate that our method outperforms state-of-the-art methods. The code will be released.

cs.CV↗
arXiv2023

Structured Light Modal Interface via Liquid-Crystal Planar Optics

Recent advances in planar optics with geometric-phase superstructures have brought a new paradigm in the control of structured light and, in particular, has substantially enhanced the capabilities of generating and detecting orbital angular momentum (OAM) states of light and associated spatial modes. However, the structured modal interface that can reciprocally link OAM states via adiabatic control and access-associated higher-order geometric phase remains absent in planar optics. In this work, we propose and experimentally demonstrate a planar optical astigmatic retarder fabricated with liquid-crystal (LC) geometric phase. The LC superstructure was designed with the principle of fractional Fourier transformation and is capable of reciprocal conversion between all possible OAM states on the same modal sphere. Such a planar device paves the way towards an easily deployed modal interface of paraxial OAM states, unlocks the resource of higher-order geometric phase, and has promising applications in high-dimensional classical/quantum information.

physics.optics↗
arXiv2023

Broad-Wavevector Spin Pumping of Flat-Band Magnons

We report the experimental observation of large spin pumping signals in YIG/Pt system driven by broad-wavevector spin-wave spin current. 280 nm-wide microwave inductive antennas offer broad-wavevector excitation which, in combination with quasi-flatband of YIG, allows a large number of magnons to participate in spin pumping at a given frequency. Through comparison with ferromagnetic resonance spin pumping, we attribute the enhancement of the spin current to the multichromatic magnons. The high efficiency of spin current generation enables us to uncover nontrivial propagating properties in ultra-low power regions. Additionally, our study achieves the spatially separated detection of magnons, allowing the direct extraction of the decay length. The synergistic combination of the capability of broad-wavevector excitation, enhanced voltage signals, and nonlocal detection provides a new avenue for the electrical exploration of spin waves dynamics.

cond-mat.mes-hall↗
arXiv2023

Ultra-Long Sequence Distributed Transformer

Transformer models trained on long sequences often achieve higher accuracy than short sequences. Unfortunately, conventional transformers struggle with long sequence training due to the overwhelming computation and memory requirements. Existing methods for long sequence training offer limited speedup and memory reduction, and may compromise accuracy. This paper presents a novel and efficient distributed training method, the Long Short-Sequence Transformer (LSS Transformer), for training transformer with long sequences. It distributes a long sequence into segments among GPUs, with each GPU computing a partial self-attention for its segment. Then, it uses a fused communication and a novel double gradient averaging technique to avoid the need to aggregate partial self-attention and minimize communication overhead. We evaluated the performance between LSS Transformer and the state-of-the-art Nvidia sequence parallelism on a Wikipedia enwik8 dataset. Results show that our proposed method lead to 5.6x faster and 10.2x more memory-efficient implementation compared to state-of-the-art sequence parallelism on 144 Nvidia V100 GPUs. Moreover, our algorithm scales to an extreme sequence length of 50,112 at 3,456 GPUs, achieving 161% super-linear parallel efficiency and a throughput of 32 petaflops.

cs.DC↗
arXiv2023

A RIS-Based Vehicle DOA Estimation Method With Integrated Sensing and Communication System

With the development of intelligent transportation, growing attention has been received to integrated sensing and communication (ISAC) systems. In this paper, we formulate a novel passive sensing technique to obtain information on the vehicle's direction of arrival (DOA) using reconfigurable intelligent surfaces (RIS). A novel estimation method is proposed in the scenario with a receiver using only one full-functional channel, where multiple measurements for the DOA estimation are achieved by controlling the reflection matrix (measurement matrix) in the RIS. Moreover, different from the existing estimation methods, we also consider the interference signals introduced by wireless communication in the ISAC system. Then, we propose a novel atomic norm-based method to remove the interference signals and reconstruct the sparse signal. Additionally, a novel Hankel-based multiple signal classification (MUSIC) method is formulated to obtain the DOA information after the interference removal. To reduce the interference signals more efficiently and improve the performance of the sparse reconstruction, we optimize the measurement matrix to improve the signal-to-interference-plus-noise ratio (SINR). Finally, the theoretical Cram'{e}r-Rao lower bound (CRLB) is derived for the ISAC system on the vehicle DOA estimation. Simulation results show that the proposed method can achieve better performance in the DOA estimation, and the corresponding CRLB with different distributions of the sensing nodes are shown. The code for the proposed method is available online https://github.com/chenpengseu/PassiveDOA-ISAC-RIS.git.

eess.SP↗
arXiv2023

A Unified Framework for Modality-Agnostic Deepfakes Detection

As AI-generated content (AIGC) thrives, deepfakes have expanded from single-modality falsification to cross-modal fake content creation, where either audio or visual components can be manipulated. While using two unimodal detectors can detect audio-visual deepfakes, cross-modal forgery clues could be overlooked. Existing multimodal deepfake detection methods typically establish correspondence between the audio and visual modalities for binary real/fake classification, and require the co-occurrence of both modalities. However, in real-world multi-modal applications, missing modality scenarios may occur where either modality is unavailable. In such cases, audio-visual detection methods are less practical than two independent unimodal methods. Consequently, the detector can not always obtain the number or type of manipulated modalities beforehand, necessitating a fake-modality-agnostic audio-visual detector. In this work, we introduce a comprehensive framework that is agnostic to fake modalities, which facilitates the identification of multimodal deepfakes and handles situations with missing modalities, regardless of the manipulations embedded in audio, video, or even cross-modal forms. To enhance the modeling of cross-modal forgery clues, we employ audio-visual speech recognition (AVSR) as a preliminary task. This efficiently extracts speech correlations across modalities, a feature challenging for deepfakes to replicate. Additionally, we propose a dual-label detection approach that follows the structure of AVSR to support the independent detection of each modality. Extensive experiments on three audio-visual datasets show that our scheme outperforms state-of-the-art detection methods with promising performance on modality-agnostic audio/video deepfakes.

cs.MM↗
arXiv2023

At the Locus of Performance: Quantifying the Effects of Copious 3D-Stacked Cache on HPC Workloads

Over the last three decades, innovations in the memory subsystem were primarily targeted at overcoming the data movement bottleneck. In this paper, we focus on a specific market trend in memory technology: 3D-stacked memory and caches. We investigate the impact of extending the on-chip memory capabilities in future HPC-focused processors, particularly by 3D-stacked SRAM. First, we propose a method oblivious to the memory subsystem to gauge the upper-bound in performance improvements when data movement costs are eliminated. Then, using the gem5 simulator, we model two variants of a hypothetical LARge Cache processor (LARC), fabricated in 1.5 nm and enriched with high-capacity 3D-stacked cache. With a volume of experiments involving a broad set of proxy-applications and benchmarks, we aim to reveal how HPC CPU performance will evolve, and conclude an average boost of 9.56x for cache-sensitive HPC applications, on a per-chip basis. Additionally, we exhaustively document our methodological exploration to motivate HPC centers to drive their own technological agenda through enhanced co-design.

cs.DC↗
arXiv2023

Derivative-Informed Neural Operator: An Efficient Framework for High-Dimensional Parametric Derivative Learning

We propose derivative-informed neural operators (DINOs), a general family of neural networks to approximate operators as infinite-dimensional mappings from input function spaces to output function spaces or quantities of interest. After discretizations both inputs and outputs are high-dimensional. We aim to approximate not only the operators with improved accuracy but also their derivatives (Jacobians) with respect to the input function-valued parameter to empower derivative-based algorithms in many applications, e.g., Bayesian inverse problems, optimization under parameter uncertainty, and optimal experimental design. The major difficulties include the computational cost of generating derivative training data and the high dimensionality of the problem leading to large training cost. To address these challenges, we exploit the intrinsic low-dimensionality of the derivatives and develop algorithms for compressing derivative information and efficiently imposing it in neural operator training yielding derivative-informed neural operators. We demonstrate that these advances can significantly reduce the costs of both data generation and training for large classes of problems (e.g., nonlinear steady state parametric PDE maps), making the costs marginal or comparable to the costs without using derivatives, and in particular independent of the discretization dimension of the input and output functions. Moreover, we show that the proposed DINO achieves significantly higher accuracy than neural operators trained without derivative information, for both function approximation and derivative approximation (e.g., Gauss-Newton Hessian), especially when the training data are limited.

math.NA↗
arXiv2023

Magneto-optical trap reaction microscope for photoionization of cold strontium atoms

We developed a magneto-optical trap reaction microscope (MOTREMI) for strontium atoms by combining the multi-particle coincident detection with laser cooling technique. Present compact injection system can provide cold Sr atoms in three modes of 2D MOT, molasses and 3D MOT, delivering targets with adjustable densities and ratios of the ground state $5s^2$ ($^1S_{0}$) and the excited states $5s5p$ ($^{1}P_{1}$ and $^{3}P_{J}$ etc). The target profiles for the temperature, the density and the size of 3D MOT as well as cold atomic flux in 2D MOT model were characterized in details. With present state-of-the-art setup, we demonstrated the single photoionization of Sr atoms with molasses by absorption of few 800-nm photons, where Sr$^+$ and $e$ were detected in coincidence and most of ionization channels were identified taking into account photoelectron energy, laser-intensity dependence, and target dependence. The best momentum resolution of coincident Sr$^+$ and $e$ along time-of-flight are achieved up to 0.12 a.u. and 0.02 a.u., respectively. Present photoelectron momentum distributions ionized from the ground state and a few excited states illuminate unprecedentedly rich landscapes manifesting prominent features for multi-photon absorption. The full vector momenta of electrons and recoil ion in coincidence paves the way to further studying two-electron correlation dynamics and multi-electron effects in the multiple ionization of alkaline-earth atoms in the ultraviolet region.

physics.atom-ph↗
arXiv2023

Approximation of the invariant measure for stable SDE by the Euler-Maruyama scheme with decreasing step-sizes

Let $(X_t)_{t \ge 0}$ be the solution of the stochastic differential equation $$dX_t = b(X_t) dt+A dZ_t, \quad X_{0}=x,$$ where $b: \mathbb{R}^d \rightarrow \mathbb R^d$ is a Lipschitz function, $A \in \mathbb R^{d \times d}$ is a positive definite matrix, $(Z_t)_{t\geq 0}$ is a $d$-dimensional rotationally invariant $α$-stable Lévy process with $α\in (1,2)$ and $x\in\mathbb{R}^{d}$. We use two Euler-Maruyama schemes with decreasing step sizes $Γ= (γ_n)_{n\in \mathbb{N}}$ to approximate the invariant measure of $(X_t)_{t \ge 0}$: one with i.i.d. $α$-stable distributed random variables as its innovations and the other with i.i.d. Pareto distributed random variables as its innovations. We study the convergence rate of these two approximation schemes in the Wasserstein-1 distance. For the first scheme, when the function $b$ is Lipschitz and satisfies a certain dissipation condition, we show that the convergence rate is $γ^{1/α}_n$. Under an additional assumption on the second order directional derivatives of $b$, this convergence rate can be improved to $γ^{1+\frac 1 α-\frac{1}κ}_n$ for any $κ\in [1,α)$. For the second scheme, when the function $b$ is twice continuously differentiable, we obtain a convergence rate of $γ^{\frac{2-α}α}_n$. We show that the rate $γ^{\frac{2-α}α}_n$ is optimal for the one dimensional stable Ornstein-Uhlenbeck process. Our theorems indicate that the recent remarkable result about the unadjusted Langevin algorithm with additive innovations can be extended to the SDEs driven by an $α$-stable Lévy process and the corresponding convergence rate has a similar behaviour. Compared with the previous result, we have relaxed the second order differentiability condition to the Lipschitz condition for the first scheme.

math.PR↗
arXiv2023

BEVHeight++: Toward Robust Visual Centric 3D Object Detection

While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird's eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight++, to address this issue. In essence, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. By incorporating both height and depth encoding techniques, we achieve a more accurate and robust projection from 2D to BEV spaces. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. In terms of the ego-vehicle scenario, our BEVHeight++ possesses superior over depth-only methods. Specifically, it yields a notable improvement of +1.9% NDS and +1.1% mAP over BEVDepth when evaluated on the nuScenes validation set. Moreover, on the nuScenes test set, our method achieves substantial advancements, with an increase of +2.8% NDS and +1.7% mAP, respectively.

cs.CV↗
arXiv2023

NoncovANM: Gridless DOA Estimation for LPDF System

Direction of arrival (DOA) estimation is an important research in the area of array signal processing, and has been studied for decades. High resolution DOA estimation requires large array aperture, which leads to the increase of hardware cost. Besides, high accuracy DOA estimation methods usually have high computational complexity. In this paper, the problem of decreasing the hardware cost and algorithm complexity is addressed. First, considering the ability of flexible controlling the electromagnetic waves and low-cost, an intelligent reconfigurable surface (IRS)-aided low-cost passive direction finding (LPDF) system is developed, where only one fully functional receiving channel is adopted. Then, the sparsity of targets direction in the spatial domain is exploited by formulating an atomic norm minimization (ANM) problem to estimate the DOA. Traditionally, solving ANM problem is complex and cannot be realized efficiently. Hence, a novel nonconvex-based ANM (NC-ANM) method is proposed by gradient threshold iteration, where a perturbation is introduced to avoid falling into saddle points. The theoretical analysis for the convergence of the NC-ANM method is also given. Moreover, the corresponding Cramér-Rao lower bound (CRLB) in the LPDF system is derived, and taken as the referred bound of the DOA estimation. Simulation results show that the proposed method outperforms the compared methods in the DOA estimation with lower computational complexity in the LPDF system.

eess.SP↗
arXiv2023

DNN-DANM: A High-Accuracy Two-Dimensional DOA Estimation Method Using Practical RIS

Reconfigurable intelligent surface (RIS) or intelligent reflecting surface (IRS) has been an attractive technology for future wireless communication and sensing systems. However, in the practical RIS, the mutual coupling effect among RIS elements, the reflection phase shift, and amplitude errors will degrade the RIS performance significantly. This paper investigates the two-dimensional direction-of-arrival (DOA) estimation problem in the scenario using a practical RIS. After formulating the system model with the mutual coupling effect and the reflection phase/amplitude errors of the RIS, a novel DNNDANM method is proposed for the DOA estimation by combining the deep neural network (DNN) and the decoupling atomic norm minimization (DANM). The DNN step reconstructs the received signal from the one with RIS impairments, and the DANM step exploits the signal sparsity in the two-dimensional spatial domain. Additionally, a semi-definite programming (SDP) method with low computational complexity is proposed to solve the atomic minimization problem. Finally, both simulation and prototype are carried out to show estimation performance, and the proposed method outperforms the existing methods in the two-dimensional DOA estimation with low complexity in the scenario with practical RIS.

eess.SP↗
arXiv2023

Flip: Data-Centric Edge CGRA Accelerator

Coarse-Grained Reconfigurable Arrays (CGRA) are promising edge accelerators due to the outstanding balance in flexibility, performance, and energy efficiency. Classic CGRAs statically map compute operations onto the processing elements (PE) and route the data dependencies among the operations through the Network-on-Chip. However, CGRAs are designed for fine-grained static instruction-level parallelism and struggle to accelerate applications with dynamic and irregular data-level parallelism, such as graph processing. To address this limitation, we present Flip, a novel accelerator that enhances traditional CGRA architectures to boost the performance of graph applications. Flip retains the classic CGRA execution model while introducing a special data-centric mode for efficient graph processing. Specifically, it exploits the natural data parallelism of graph algorithms by mapping graph vertices onto processing elements (PEs) rather than the operations, and supporting dynamic routing of temporary data according to the runtime evolution of the graph frontier. Experimental results demonstrate that Flip achieves up to 36$\times$ speedup with merely 19% more area compared to classic CGRAs. Compared to state-of-the-art large-scale graph processors, Flip has similar energy efficiency and 2.2$\times$ better area efficiency at a much-reduced power/area budget.

cs.AR↗
arXiv2023

HOPPER: Interpretative Fuzzing for Libraries

Despite the fact that the state-of-the-art fuzzers can generate inputs efficiently, existing fuzz drivers still cannot adequately cover entries in libraries. Most of these fuzz drivers are crafted manually by developers, and their quality depends on the developers' understanding of the code. Existing works have attempted to automate the generation of fuzz drivers by learning API usage from code and execution traces. However, the generated fuzz drivers are limited to a few specific call sequences by the code being learned. To address these challenges, we present HOPPER, which can fuzz libraries without requiring any domain knowledge to craft fuzz drivers. It transforms the problem of library fuzzing into the problem of interpreter fuzzing. The interpreters linked against libraries under test can interpret the inputs that describe arbitrary API usage. To generate semantically correct inputs for the interpreter, HOPPER learns the intra- and inter-API constraints in the libraries and mutates the program with grammar awareness. We implemented HOPPER and evaluated its effectiveness on 11 real-world libraries against manually crafted fuzzers and other automatic solutions. Our results show that HOPPER greatly outperformed the other fuzzers in both code coverage and bug finding, having uncovered 25 previously unknown bugs that other fuzzers couldn't. Moreover, we have demonstrated that the proposed intra- and inter-API constraint learning methods can correctly learn constraints implied by the library and, therefore, significantly improve the fuzzing efficiency. The experiment results indicate that HOPPER is able to explore a vast range of API usages for library fuzzing out of the box.

cs.CR↗
arXiv2023

Towards General and Efficient Online Tuning for Spark

The distributed data analytic system -- Spark is a common choice for processing massive volumes of heterogeneous data, while it is challenging to tune its parameters to achieve high performance. Recent studies try to employ auto-tuning techniques to solve this problem but suffer from three issues: limited functionality, high overhead, and inefficient search. In this paper, we present a general and efficient Spark tuning framework that can deal with the three issues simultaneously. First, we introduce a generalized tuning formulation, which can support multiple tuning goals and constraints conveniently, and a Bayesian optimization (BO) based solution to solve this generalized optimization problem. Second, to avoid high overhead from additional offline evaluations in existing methods, we propose to tune parameters along with the actual periodic executions of each job (i.e., online evaluations). To ensure safety during online job executions, we design a safe configuration acquisition method that models the safe region. Finally, three innovative techniques are leveraged to further accelerate the search process: adaptive sub-space generation, approximate gradient descent, and meta-learning method. We have implemented this framework as an independent cloud service, and applied it to the data platform in Tencent. The empirical results on both public benchmarks and large-scale production tasks demonstrate its superiority in terms of practicality, generality, and efficiency. Notably, this service saves an average of 57.00% memory cost and 34.93% CPU cost on 25K in-production tasks within 20 iterations, respectively.

cs.DC↗