arXiv subjects

Yuxuan Li

Publications and source records attributed to Yuxuan Li.

At least 145 records · Page 8Linked to original sources

arXiv2024

Deep Adversarial Defense Against Multilevel-Lp Attacks

Deep learning models have shown considerable vulnerability to adversarial attacks, particularly as attacker strategies become more sophisticated. While traditional adversarial training (AT) techniques offer some resilience, they often focus on defending against a single type of attack, e.g., the $\ell_\infty$-norm attack, which can fail for other types. This paper introduces a computationally efficient multilevel $\ell_p$ defense, called the Efficient Robust Mode Connectivity (EMRC) method, which aims to enhance a deep learning model's resilience against multiple $\ell_p$-norm attacks. Similar to analytical continuation approaches used in continuous optimization, the method blends two $p$-specific adversarially optimal models, the $\ell_1$- and $\ell_\infty$-norm AT solutions, to provide good adversarial robustness for a range of $p$. We present experiments demonstrating that our approach performs better on various attacks as compared to AT-$\ell_\infty$, E-AT, and MSD, for datasets/architectures including: CIFAR-10, CIFAR-100 / PreResNet110, WideResNet, ViT-Base.

cs.LG↗

arXiv2024

Observation of Heat Pumping Effect by Radiative Shuttling

Heat shuttling phenomenon is characterized by the presence of a non-zero heat flow between two bodies without net thermal bias on average. It was initially predicted in the context of nonlinear heat conduction within atomic lattices coupled to two time-oscillating thermostats. Recent theoretical works revealed an analog of this effect for heat exchanges mediated by thermal photons between two solids having a temperature dependent emissivity. In this paper, we present the experimental proof of this effect using systems made with composite materials based on phase change materials. By periodically modulating the temperature of one of two solids we report that the system akin to heat pumping with a controllable heat flow direction. Additionally, we demonstrate the effectiveness of a simultaneous modulation of two temperatures to control both the strength and direction of heat shuttling by exploiting the phase delay between these temperatures. These results show that this effect is promising for an active thermal management of solid-state technology, to cool down solids, to insulate them from their background or to amplify heat exchanges.

physics.app-ph↗

arXiv2024

General Scintillation for Gaussian Beam Propagating through Oceanic Turbulence and UWOC System Performance Evaluation

In this paper, we derive a general and exact closed-form expression of scintillation index (SI) for a Gaussian beam propagating through weak oceanic turbulence, based on the general oceanic turbulence optical power spectrum (OTOPS) and the Rytov theory. Our universal expression not only includes existing Rytov variances but also accounts for actual cases where the Kolmogorov microscale is non-zero. The correctness and accuracy of our derivation are verified through comparison with the published work under identical conditions. By utilizing our derived expressions, we analyze the impact of various beam, propagation and oceanic turbulence parameters on both SI and bit error rate (BER) performance of underwater wireless optical communication (UWOC) systems. Numerical results demonstrate that the relationship between the Kolmogorov microscale and SI is nonlinear. Additionally, considering that certain oceanic turbulence parameters are related to depth, we use temperature and salinity data from Argo buoy deployed in real oceans to investigate the dependence of SI on depth. Our findings will contribute to the design and optimization of UWOC systems.

physics.optics↗

arXiv2024

Transient Measurement of Near-field Thermal Radiation between Macroscopic Objects

The involvement of evanescent waves in the near-field regime could greatly enhance the spontaneous thermal radiation, offering a unique opportunity to study nanoscale photon-phonon interaction. However, accurately characterizing this subtle phenomenon is very challenging. This paper proposes a transient all-optical method for rapidly characterizing near-field radiative heat transfer (NFRHT) between macroscopic objects, using the first law of thermodynamics. Significantly, a full measurement at a fixed gap distance is completed within tens of seconds. By simplifying the configuration, the transient all-optical method achieves high measurement accuracy and reliable reproducibility. The proposed method can effectively analyze the NFRHT in various material systems, including SiO2, SiC, and Si, which involve different phonon or plasmon polaritons. Experimental observations demonstrate significant super-Planckian radiation, which arises from the near-field coupling of bounded surface modes. Furthermore, the method achieves excellent agreement with theory, with a minimal discrepancy of less than 2.7% across a wide temperature range. This wireless method could accurately characterize the NFRHT for objects with different sizes or optical properties, enabling the exploration of both fundamental interests and practical applications.

physics.optics↗

arXiv2024

Radiative Thermal Transistor

Developing thermal analogues of field-effect transistor could open the door to a low-power and even zero-power communication technology working with heat rather than electricity. These solid-sate devices could also find many applications in the field of active thermal management in numerous technologies (microelectronic, building science, energy harvesting,conversion,...). Recent theoretical works has suggested that a photonic transistor made with three terminals can in principle be used to switch, modulate, and even amplify heat flux through exchange of thermal photons. Here, we report an experimental demonstration of thermal transistor effect using a non-contact system composed by a temperature-controlled metal-insulator-based material interacting in far-field regime with two blackbodies held at two different temperatures. We demonstrate that, with a tiny change in the temperature of the active layer, the heat flux received by the cold blackbody can be drastically modified. An amplification parameter of heat flux over 20 is reported.

physics.app-ph↗

arXiv2024

A Human-Computer Collaborative Tool for Training a Single Large Language Model Agent into a Network through Few Examples

The capabilities of a single large language model (LLM) agent for solving a complex task are limited. Connecting multiple LLM agents to a network can effectively improve overall performance. However, building an LLM agent network (LAN) requires a substantial amount of time and effort. In this paper, we introduce EasyLAN, a human-computer collaborative tool that helps developers construct LANs. EasyLAN initially generates a LAN containing only one agent based on the description of the desired task. Subsequently, EasyLAN leverages a few training examples to update the LAN. For each example, EasyLAN models the gap between the output and the ground truth and identifies the causes of the errors. These errors are addressed through carefully designed strategies. Users can intervene in EasyLAN's workflow or directly modify the LAN. Eventually, the LAN evolves from a single agent to a network of LLM agents. The experimental results indicate that developers can rapidly construct LANs with good performance.

cs.HC↗

arXiv2024

Kilometer-Level Coupled Modeling Using 40 Million Cores: An Eight-Year Journey of Model Development

With current and future leading systems adopting heterogeneous architectures, adapting existing models for heterogeneous supercomputers is of urgent need for improving model resolution and reducing modeling uncertainty. This paper presents our three-week effort on porting a complex earth system model, CESM 2.2, to a 40-million-core Sunway supercomputer. Taking a non-intrusive approach that tries to minimizes manual code modifications, our project tries to achieve both improvement of performance and consistency of the model code. By using a hierarchical grid system and an OpenMP-based offloading toolkit, our porting and parallelization effort covers over 80% of the code, and achieves a simulation speed of 340 SDPD (simulated days per day) for 5-km atmosphere, 265 SDPD for 3-km ocean, and 222 SDPD for a coupled model, thus making multi-year or even multi-decadal experiments at such high resolution possible.

cs.DC↗

arXiv2024

ADs: Active Data-sharing for Data Quality Assurance in Advanced Manufacturing Systems

Machine learning (ML) methods are widely used in industrial applications, which usually require a large amount of training data. However, data collection needs extensive time costs and investments in the manufacturing system, and data scarcity commonly exists. Therefore, data-sharing is widely enabled among multiple machines with similar functionality to augment the dataset for building ML methods. However, distribution mismatch inevitably exists in their data due to different working conditions, while the ML methods are assumed to be built and tested on the dataset following the same distribution. Thus, an Active Data-sharing (ADs) framework is proposed to ensure the quality of the shared data among multiple machines. It is designed to simultaneously select the most informative data points benefiting the downstream tasks and mitigate the distribution mismatch among all selected data points. The proposed method is validated on anomaly detection on in-situ monitoring data from three additive manufacturing processes.

cs.LG↗

arXiv2024

Attention-stacked Generative Adversarial Network (AS-GAN)-empowered Sensor Data Augmentation for Online Monitoring of Manufacturing System

Machine learning (ML) has been extensively adopted for the online sensing-based monitoring in advanced manufacturing systems. However, the sensor data collected under abnormal states are usually insufficient, leading to significant data imbalanced issue for supervised machine learning. A common solution is to incorporate data augmentation techniques, i.e., augmenting the available abnormal states data (i.e., minority samples) via synthetic generation. To generate the high-quality minority samples, it is vital to learn the underlying distribution of the abnormal states data. In recent years, the generative adversarial network (GAN)-based approaches become popular to learn data distribution as well as perform data augmentation. However, in practice, the quality of generated samples from GAN-based data augmentation may vary drastically. In addition, the sensor signals are collected sequentially by time from the manufacturing systems, which means sequential information is also very important in data augmentation. To address these limitations, inspired by the multi-head attention mechanism, this paper proposed an attention-stacked GAN (AS-GAN) architecture for sensor data augmentation of online monitoring in manufacturing system. It incorporates a new attention-stacked framework to strengthen the generator in GAN with the capability of capturing sequential information, and thereby the developed attention-stacked framework greatly helps to improve the quality of the generated sensor signals. Afterwards, the generated high-quality sensor signals for abnormal states could be applied to train classifiers more accurately, further improving the online monitoring performance of manufacturing systems. The case study conducted in additive manufacturing also successfully validated the effectiveness of the proposed AS-GAN.

cs.LG↗

arXiv2024

SpectralGPT: Spectral Remote Sensing Foundation Model

The foundation model has recently garnered significant attention due to its potential to revolutionize the field of visual representation learning in a self-supervised manner. While most foundation models are tailored to effectively process RGB images for various visual tasks, there is a noticeable gap in research focused on spectral data, which offers valuable information for scene understanding, especially in remote sensing (RS) applications. To fill this gap, we created for the first time a universal RS foundation model, named SpectralGPT, which is purpose-built to handle spectral RS images using a novel 3D generative pretrained transformer (GPT). Compared to existing foundation models, SpectralGPT 1) accommodates input images with varying sizes, resolutions, time series, and regions in a progressive training fashion, enabling full utilization of extensive RS big data; 2) leverages 3D token generation for spatial-spectral coupling; 3) captures spectrally sequential patterns via multi-target reconstruction; 4) trains on one million spectral RS images, yielding models with over 600 million parameters. Our evaluation highlights significant performance improvements with pretrained SpectralGPT models, signifying substantial potential in advancing spectral RS big data applications within the field of geoscience across four downstream tasks: single/multi-label scene classification, semantic segmentation, and change detection.

cs.CV↗

arXiv2024

The second largest eigenvalue of some nonnormal Cayley graphs on symmetric groups

A Cayley graph on the symmetric group $S_n$ is said to have the Aldous property if its strictly second largest eigenvalue (that is, the largest eigenvalue strictly smaller than the degree) is attained by the standard representation of $S_n$. For $1\leq r < k < n$, let $C(n,k;r)$ be the set of $k$-cycles of $S_n$ moving every point in $\{1, \ldots, r\}$. Recently, Siemons and Zalesski [J. Algebraic Combin. 55 (2022) 989--1005] posed a conjecture which is equivalent to saying that for any $n \ge 5$ and $1\leq r<k<n$ the nonnormal Cayley graph $\mathrm{Cay}(S_n, C(n,k;r))$ on $S_n$ with connection set $C(n,k;r)$ has the Aldous property. Solving this conjecture, we prove that all these graphs have the Aldous property except when (i) $(n, k, r) = (6, 5, 1)$ or (ii) $n$ is odd, $k = n-1$, and $1 \le r < \frac{n}{2}$. Along the way we determine all irreducible representations of $S_n$ that can achieve the strictly second largest eigenvalue of $\mathrm{Cay}(S_n, C(n,n-1;r))$ as well as the smallest eigenvalue of this graph.

math.CO↗

arXiv2024

Model-Informed Generative Adversarial Network (MI-GAN) for Learning Optimal Power Flow

The optimal power flow (OPF) problem, as a critical component of power system operations, becomes increasingly difficult to solve due to the variability, intermittency, and unpredictability of renewable energy brought to the power system. Although traditional optimization techniques, such as stochastic and robust optimization approaches, could be leveraged to address the OPF problem, in the face of renewable energy uncertainty, i.e., the dynamic coefficients in the optimization model, their effectiveness in dealing with large-scale problems remains limited. As a result, deep learning techniques, such as neural networks, have recently been developed to improve computational efficiency in solving OPF problems with the utilization of data. However, the feasibility and optimality of the solution may not be guaranteed, and the system dynamics cannot be properly addressed as well. In this paper, we propose an optimization model-informed generative adversarial network (MI-GAN) framework to solve OPF under uncertainty. The main contributions are summarized into three aspects: (1) to ensure feasibility and improve optimality of generated solutions, three important layers are proposed: feasibility filter layer, comparison layer, and gradient-guided layer; (2) in the GAN-based framework, an efficient model-informed selector incorporating these three new layers is established; and (3) a new recursive iteration algorithm is also proposed to improve solution optimality and handle the system dynamics. The numerical results on IEEE test systems show that the proposed method is very effective and promising.

cs.LG↗

arXiv2023

Understanding Representations Pretrained with Auxiliary Losses for Embodied Agent Planning

Pretrained representations from large-scale vision models have boosted the performance of downstream embodied policy learning. We look to understand whether additional self-supervised pretraining on exploration trajectories can build on these general-purpose visual representations to better support embodied planning in realistic environments. We evaluated four common auxiliary losses in embodied AI, two hindsight-based losses, and a standard imitation learning loss, by pretraining the agent's visual compression module and state belief representations with each objective and using CLIP as a representative visual backbone. The learned representations are then frozen for downstream multi-step evaluation on two goal-directed tasks. Surprisingly, we find that imitation learning on these exploration trajectories out-performs all other auxiliary losses even despite the exploration trajectories being dissimilar from the downstream tasks. This suggests that imitation of exploration may be ''all you need'' for building powerful planning representations. Additionally, we find that popular auxiliary losses can benefit from simple modifications to improve their support for downstream planning ability.

cs.RO↗

arXiv2023

Cross-City Matters: A Multimodal Remote Sensing Benchmark Dataset for Cross-City Semantic Segmentation using High-Resolution Domain Adaptation Networks

Artificial intelligence (AI) approaches nowadays have gained remarkable success in single-modality-dominated remote sensing (RS) applications, especially with an emphasis on individual urban environments (e.g., single cities or regions). Yet these AI models tend to meet the performance bottleneck in the case studies across cities or regions, due to the lack of diverse RS information and cutting-edge solutions with high generalization ability. To this end, we build a new set of multimodal remote sensing benchmark datasets (including hyperspectral, multispectral, SAR) for the study purpose of the cross-city semantic segmentation task (called C2Seg dataset), which consists of two cross-city scenes, i.e., Berlin-Augsburg (in Germany) and Beijing-Wuhan (in China). Beyond the single city, we propose a high-resolution domain adaptation network, HighDAN for short, to promote the AI model's generalization ability from the multi-city environments. HighDAN is capable of retaining the spatially topological structure of the studied urban scene well in a parallel high-to-low resolution fusion fashion but also closing the gap derived from enormous differences of RS image representations between different cities by means of adversarial learning. In addition, the Dice loss is considered in HighDAN to alleviate the class imbalance issue caused by factors across cities. Extensive experiments conducted on the C2Seg dataset show the superiority of our HighDAN in terms of segmentation performance and generalization ability, compared to state-of-the-art competitors. The C2Seg dataset and the semantic segmentation toolbox (involving the proposed HighDAN) will be available publicly at https://github.com/danfenghong.

cs.CV↗

arXiv2023

Towards Efficient Control Flow Handling in Spatial Architecture via Architecting the Control Flow Plane

Spatial architecture is a high-performance architecture that uses control flow graphs and data flow graphs as the computational model and producer/consumer models as the execution models. However, existing spatial architectures suffer from control flow handling challenges. Upon categorizing their PE execution models, we find that they lack autonomous, peer-to-peer, and temporally loosely-coupled control flow handling capability. This leads to limited performance in intensive control programs. A spatial architecture, Marionette, is proposed, with an explicit-designed control flow plane. The Control Flow Plane enables autonomous, peer-to-peer and temporally loosely-coupled control flow handling. The Proactive PE Configuration ensures timely and computation-overlapped configuration to improve handling Branch Divergence. The Agile PE Assignment enhance the pipeline performance of Imperfect Loops. We develop full stack of Marionette (ISA, compiler, simulator, RTL) and demonstrate that in a variety of challenging intensive control programs, compared to state-of-the-art spatial architectures, Marionette outperforms Softbrain, TIA, REVEL, and RipTide by geomean 2.88x, 3.38x, 1.55x, and 2.66x.

cs.AR↗

arXiv2023

A solution to Babai's problem on digraphs with non-diagonalizable adjacency matrix

The fact that the adjacency matrix of every finite graph is diagonalizable plays a fundamental role in spectral graph theory. Since this fact does not hold in general for digraphs, it is natural to ask whether it holds for digraphs with certain level of symmetry. Interest on this question dates back to early 1980s, when P.~J.~Cameron asked for the existence of arc-transitive digraphs with non-diagonalizable adjacency matrix. This was answered in the affirmative by L.~Babai in 1985. Then Babai posed the open problem of constructing a 2-arc-transitive digraph and a vertex-primitive digraph whose adjacency matrices are not diagonalizable. In this paper, we solve Babai's problem by constructing an infinite family of $s$-arc-transitive digraphs for each integer $s\geq2$, and an infinite family of vertex-primitive digraphs, respectively, both of whose adjacency matrices are non-diagonalizable.

math.CO↗

arXiv2023

O2ATH: An OpenMP Offloading Toolkit for the Sunway Heterogeneous Manycore Platform

The next generation Sunway supercomputer employs the SW26010pro processor, which features a specialized on-chip heterogeneous architecture. Applications with significant hotspots can benefit from the great computation capacity improvement of Sunway many-core architectures by carefully making intensive manual many-core parallelization efforts. However, some legacy projects with large codebases, such as CESM, ROMS and WRF, contain numerous lines of code and do not have significant hotspots. The cost of manually porting such applications to the Sunway architecture is almost unaffordable. To overcome such a challenge, we have developed a toolkit named O2ATH. O2ATH forwards GNU OpenMP runtime library calls to Sunway's Athread library, which greatly simplifies the parallelization work on the Sunway architecture.O2ATH enables users to write both MPE and CPE code in a single file, and parallelization can be achieved by utilizing OpenMP directives and attributes. In practice, O2ATH has helped us to port two large projects, CESM and ROMS, to the CPEs of the next generation Sunway supercomputers via the OpenMP offload method. In the experiments, kernel speedups range from 3 to 15 times, resulting in 3 to 6 times whole application speedups.Furthermore, O2ATH requires significantly fewer code modifications compared to manually crafting CPE functions.This indicates that O2ATH can greatly enhance development efficiency when porting or optimizing large software projects on Sunway supercomputers.

cs.PL↗

arXiv2023

Gaussian Boson Sampling with Pseudo-Photon-Number Resolving Detectors and Quantum Computational Advantage

We report new Gaussian boson sampling experiments with pseudo-photon-number-resolving detection, which register up to 255 photon-click events. We consider partial photon distinguishability and develop a more complete model for the characterization of the noisy Gaussian boson sampling. In the quantum computational advantage regime, we use Bayesian tests and correlation function analysis to validate the samples against all current classical mockups. Estimating with the best classical algorithms to date, generating a single ideal sample from the same distribution on the supercomputer Frontier would take ~ 600 years using exact methods, whereas our quantum computer, Jiuzhang 3.0, takes only 1.27 us to produce a sample. Generating the hardest sample from the experiment using an exact algorithm would take Frontier ~ 3.1*10^10 years.

quant-ph↗