Searcharxiv⌕ Search

arXiv subjects

Rui Zhu

Publications and source records attributed to Rui Zhu.

At least 91 records · Page 5Linked to original sources
arXiv2024

The KnowWhereGraph Ontology

KnowWhereGraph is one of the largest fully publicly available geospatial knowledge graphs. It includes data from 30 layers on natural hazards (e.g., hurricanes, wildfires), climate variables (e.g., air temperature, precipitation), soil properties, crop and land-cover types, demographics, and human health, various place and region identifiers, among other themes. These have been leveraged through the graph by a variety of applications to address challenges in food security and agricultural supply chains; sustainability related to soil conservation practices and farm labor; and delivery of emergency humanitarian aid following a disaster. In this paper, we introduce the ontology that acts as the schema for KnowWhereGraph. This broad overview provides insight into the requirements and design specifications for the graph and its schema, including the development methodology (modular ontology modeling) and the resources utilized to implement, materialize, and deploy KnowWhereGraph with its end-user interfaces and public query SPARQL endpoint.

cs.AI↗
arXiv2024

Feature Augmentation for Self-supervised Contrastive Learning: A Closer Look

Self-supervised contrastive learning heavily relies on the view variance brought by data augmentation, so that it can learn a view-invariant pre-trained representation. Beyond increasing the view variance for contrast, this work focuses on improving the diversity of training data, to improve the generalization and robustness of the pre-trained models. To this end, we propose a unified framework to conduct data augmentation in the feature space, known as feature augmentation. This strategy is domain-agnostic, which augments similar features to the original ones and thus improves the data diversity. We perform a systematic investigation of various feature augmentation architectures, the gradient-flow skill, and the relationship between feature augmentation and traditional data augmentation. Our study reveals some practical principles for feature augmentation in self-contrastive learning. By integrating feature augmentation on the instance discrimination or the instance similarity paradigm, we consistently improve the performance of pre-trained feature learning and gain better generalization over the downstream image classification and object detection task.

cs.CV↗
arXiv2024

HL-LHC sensitivity to higgsinos from natural SUSY with gravitino LSP

In the realm of natural supersymmetric models, higgsinos are typically the lightest electroweakinos. In gauge-mediated supersymmetry breaking models, the lightest higgsino-dominated particles decay into a $Z$-boson or a Higgs boson ($h$), along with an ultra-light gravitino ($\tilde{G}$) serving as the lightest supersymmetric particle (LSP). This scenario suggests a significant non-resonant $hh$ production. Basing on the recent global fitting results of the $\tilde{G}$-EWMSSM (MSSM with light electroweakinos and an eV-scale gravitino as the LSP) performed by the \textsf{GAMBIT} collaboration, which support a higgsino dominated electroweakino as light as 140 GeV, we develop two simplified models to evaluate their detection potential at the high-luminosity LHC (HL-LHC) with $\sqrt{s} = 14~{\rm TeV}$ and an integrated luminosity of $3000~{\rm fb}^{-1}$. The first model examines the processes where heavier higgsino-dominated states decay into soft $W/Z$ bosons, while the second focuses on direct decays of all three higgsino-dominated electroweakinos into $W/Z/h$ plus a $\tilde{G}$. Our study, incorporating both models and their distinct decay channels, utilizes detailed Monte Carlo simulations for signals and standard model backgrounds. We find that the HL-LHC can probe higgsinos up to 575 GeV, potentially discovering or excluding the natural SUSY scenario in the context of a gravitino LSP. Further, we reinterpret this discovery potential using the GAMBIT global fit samples, and find that the entire parameter space of $|μ| \leq 500~{\rm GeV}$ with an electroweak fine-tuning measure ($Δ_{\rm EW}$) under 100 in $\tilde{G}$-EWMSSM is accessible at the HL-LHC.

hep-ph↗
arXiv2024

Theoretical bounds on dark Higgs mass in a self-interacting dark matter model with $U(1)'$

Motivated by the null results of current dark matter searches and the small-scale problems, we study a dark sector charged by a spontaneous broken gauge $U(1)'$. To explore the parameter space of this model, in addition to the consideration of the small-scale data, we also consider the theoretical bounds on the dark Higgs mass, with the upper bound coming from the tree-level perturbative unitarity and the lower bound from the one-loop Linde-Weinberg bound. We deeply examine the dependence of the Linde-Weinberg bound on gauge choice and energy scale, and present a Linde-Weinberg bound that is gauge and scale independent. Combining the theoretical and observational constraints, we obtain the following ranges for the parameter space: the dark matter mass is 10-500 GeV, the mediator (dark photon) mass is 0.5-5 MeV, the dark Higgs mass is 0.05-50 MeV, and the dark fine-structure constant is 0.001-0.4. We conclude that the dark Higgs in this model cannot be ignored in the phenomenological study of the dark sector.

hep-ph↗
arXiv2024

The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks

The rapid advancements of large language models (LLMs) have raised public concerns about the privacy leakage of personally identifiable information (PII) within their extensive training datasets. Recent studies have demonstrated that an adversary could extract highly sensitive privacy data from the training data of LLMs with carefully designed prompts. However, these attacks suffer from the model's tendency to hallucinate and catastrophic forgetting (CF) in the pre-training stage, rendering the veracity of divulged PIIs negligible. In our research, we propose a novel attack, Janus, which exploits the fine-tuning interface to recover forgotten PIIs from the pre-training data in LLMs. We formalize the privacy leakage problem in LLMs and explain why forgotten PIIs can be recovered through empirical analysis on open-source language models. Based upon these insights, we evaluate the performance of Janus on both open-source language models and two latest LLMs, i.e., GPT-3.5-Turbo and LLaMA-2-7b. Our experiment results show that Janus amplifies the privacy risks by over 10 times in comparison with the baseline and significantly outperforms the state-of-the-art privacy extraction attacks including prefix attacks and in-context learning (ICL). Furthermore, our analysis validates that existing fine-tuning APIs provided by OpenAI and Azure AI Studio are susceptible to our Janus attack, allowing an adversary to conduct such an attack at a low cost.

cs.CR↗
arXiv2024

Selective Amnesia: On Efficient, High-Fidelity and Blind Suppression of Backdoor Effects in Trojaned Machine Learning Models

In this paper, we present a simple yet surprisingly effective technique to induce "selective amnesia" on a backdoored model. Our approach, called SEAM, has been inspired by the problem of catastrophic forgetting (CF), a long standing issue in continual learning. Our idea is to retrain a given DNN model on randomly labeled clean data, to induce a CF on the model, leading to a sudden forget on both primary and backdoor tasks; then we recover the primary task by retraining the randomized model on correctly labeled clean data. We analyzed SEAM by modeling the unlearning process as continual learning and further approximating a DNN using Neural Tangent Kernel for measuring CF. Our analysis shows that our random-labeling approach actually maximizes the CF on an unknown backdoor in the absence of triggered inputs, and also preserves some feature extraction in the network to enable a fast revival of the primary task. We further evaluated SEAM on both image processing and Natural Language Processing tasks, under both data contamination and training manipulation attacks, over thousands of models either trained on popular image datasets or provided by the TrojAI competition. Our experiments show that SEAM vastly outperforms the state-of-the-art unlearning techniques, achieving a high Fidelity (measuring the gap between the accuracy of the primary task and that of the backdoor) within a few minutes (about 30 times faster than training a model from scratch using the MNIST dataset), with only a small amount of clean data (0.1% of training data for TrojAI models).

cs.LG↗
arXiv2024

Bayesian Semi-supervised Multi-category Classification under Nonparanormality

Semi-supervised learning is a model training method that uses both labeled and unlabeled data. This paper proposes a fully Bayes semi-supervised learning algorithm that can be applied to any multi-category classification problem. We assume the labels are missing at random when using unlabeled data in a semi-supervised setting. Suppose we have $K$ classes in the data. We assume that the observations follow $K$ multivariate normal distributions depending on their true class labels after some common unknown transformation is applied to each component of the observation vector. The function is expanded in a B-splines series, and a prior is added to the coefficients. We consider a normal prior on the coefficients and constrain the values to meet the normality and identifiability constraints requirement. The precision matrices of the Gaussian distributions are given a conjugate Wishart prior, while the means are given the improper uniform prior. The resulting posterior is still conditionally conjugate, and the Gibbs sampler aided by a data-augmentation technique can thus be adopted. An extensive simulation study compares the proposed method with several other available methods. The proposed method is also applied to real datasets on diagnosing breast cancer and classification of signals. We conclude that the proposed method has a better prediction accuracy in various cases.

stat.ML↗
arXiv2024

Enhancing Security in Millimeter Wave SWIPT Networks

Millimeter wave (mmWave) communication encounters a major issue of extremely high power consumption. To address this problem, the simultaneous wireless information and power transfer (SWIPT) could be a promising technology. The mmWave frequencies are more appropriate for the SWIPT comparing to current low-frequency wireless transmissions, since mmWave base stations (BSs) can pack with large antenna arrays to achieve significant array gains and high-speed short-distance transmissions. Unfortunately, the implementation of SWIPT in the wireless communication may lead to an expanded defencelessness against the eavesdropping due to high transmission power and data spillage. It is conventionally believed that narrow beam offers inherent information-theoretic security against the eavesdropping, because only the eavesdroppers, which rely on the line-of-sight path between the legitimate transmitter and receiver, can receive strong enough signals. However, some mmWave experiments have shown that even by using highly directional mmWaves, the reflection signals caused by objects in the environment can be beneficial to the eavesdroppers. This paper studies the security performance in general mmWave SWIPT networks, and investigates the probability of successful eavesdropping under different attack models. Analytical expressions of eavesdropping success probability (ESP) of both independent and colluding eavesdroppers are derived by incorporating the random reflection paths in the environment. Theoretical analysis and simulation results reveal the effects of some key parameters on the ESP, such as the time switching strategy in SWIPT, densities of mmWave BSs, and carriers frequencies, etc. Based on the numerical and simulation results, some design suggestions of mmWave SWIPT are provided to defend against eavesdropping attacks and achieve secure communication in practice.

cs.CR↗
arXiv2024

PUAL: A Classifier on Trifurcate Positive-Unlabeled Data

Positive-unlabeled (PU) learning aims to train a classifier using the data containing only labeled-positive instances and unlabeled instances. However, existing PU learning methods are generally hard to achieve satisfactory performance on trifurcate data, where the positive instances distribute on both sides of the negative instances. To address this issue, firstly we propose a PU classifier with asymmetric loss (PUAL), by introducing a structure of asymmetric loss on positive instances into the objective function of the global and local learning classifier. Then we develop a kernel-based algorithm to enable PUAL to obtain non-linear decision boundary. We show that, through experiments on both simulated and real-world datasets, PUAL can achieve satisfactory classification on trifurcate data.

stat.ML↗
arXiv2024

Geospatial Knowledge Graphs

Geospatial knowledge graphs have emerged as a novel paradigm for representing and reasoning over geospatial information. In this framework, entities such as places, people, events, and observations are depicted as nodes, while their relationships are represented as edges. This graph-based data format lays the foundation for creating a "FAIR" (Findable, Accessible, Interoperable, and Reusable) environment, facilitating the management and analysis of geographic information. This entry first introduces key concepts in knowledge graphs along with their associated standardization and tools. It then delves into the application of knowledge graphs in geography and environmental sciences, emphasizing their role in bridging symbolic and subsymbolic GeoAI to address cross-disciplinary geospatial challenges. At the end, new research directions related to geospatial knowledge graphs are outlined.

cs.AI↗
arXiv2024

A Theoretical Understanding of Gradient Bias in Meta-Reinforcement Learning

Gradient-based Meta-RL (GMRL) refers to methods that maintain two-level optimisation procedures wherein the outer-loop meta-learner guides the inner-loop gradient-based reinforcement learner to achieve fast adaptations. In this paper, we develop a unified framework that describes variations of GMRL algorithms and points out that existing stochastic meta-gradient estimators adopted by GMRL are actually \textbf{biased}. Such meta-gradient bias comes from two sources: 1) the compositional bias incurred by the two-level problem structure, which has an upper bound of $\mathcal{O}\big(Kα^{K}\hatσ_{\text{In}}|τ|^{-0.5}\big)$ \emph{w.r.t.} inner-loop update step $K$, learning rate $α$, estimate variance $\hatσ^{2}_{\text{In}}$ and sample size $|τ|$, and 2) the multi-step Hessian estimation bias $\hatΔ_{H}$ due to the use of autodiff, which has a polynomial impact $\mathcal{O}\big((K-1)(\hatΔ_{H})^{K-1}\big)$ on the meta-gradient bias. We study tabular MDPs empirically and offer quantitative evidence that testifies our theoretical findings on existing stochastic meta-gradient estimators. Furthermore, we conduct experiments on Iterated Prisoner's Dilemma and Atari games to show how other methods such as off-policy learning and low-bias estimator can help fix the gradient bias for GMRL algorithms in general.

cs.LG↗
arXiv2024

SD-DiT: Unleashing the Power of Self-supervised Discrimination in Diffusion Transformer

Diffusion Transformer (DiT) has emerged as the new trend of generative diffusion models on image generation. In view of extremely slow convergence in typical DiT, recent breakthroughs have been driven by mask strategy that significantly improves the training efficiency of DiT with additional intra-image contextual learning. Despite this progress, mask strategy still suffers from two inherent limitations: (a) training-inference discrepancy and (b) fuzzy relations between mask reconstruction & generative diffusion process, resulting in sub-optimal training of DiT. In this work, we address these limitations by novelly unleashing the self-supervised discrimination knowledge to boost DiT training. Technically, we frame our DiT in a teacher-student manner. The teacher-student discriminative pairs are built on the diffusion noises along the same Probability Flow Ordinary Differential Equation (PF-ODE). Instead of applying mask reconstruction loss over both DiT encoder and decoder, we decouple DiT encoder and decoder to separately tackle discriminative and generative objectives. In particular, by encoding discriminative pairs with student and teacher DiT encoders, a new discriminative loss is designed to encourage the inter-image alignment in the self-supervised embedding space. After that, student samples are fed into student DiT decoder to perform the typical generative diffusion task. Extensive experiments are conducted on ImageNet dataset, and our method achieves a competitive balance between training cost and generative capacity.

cs.CV↗
arXiv2024

DPAdapter: Improving Differentially Private Deep Learning through Noise Tolerance Pre-training

Recent developments have underscored the critical role of \textit{differential privacy} (DP) in safeguarding individual data for training machine learning models. However, integrating DP oftentimes incurs significant model performance degradation due to the perturbation introduced into the training process, presenting a formidable challenge in the {differentially private machine learning} (DPML) field. To this end, several mitigative efforts have been proposed, typically revolving around formulating new DPML algorithms or relaxing DP definitions to harmonize with distinct contexts. In spite of these initiatives, the diminishment induced by DP on models, particularly large-scale models, remains substantial and thus, necessitates an innovative solution that adeptly circumnavigates the consequential impairment of model utility. In response, we introduce DPAdapter, a pioneering technique designed to amplify the model performance of DPML algorithms by enhancing parameter robustness. The fundamental intuition behind this strategy is that models with robust parameters are inherently more resistant to the noise introduced by DP, thereby retaining better performance despite the perturbations. DPAdapter modifies and enhances the sharpness-aware minimization (SAM) technique, utilizing a two-batch strategy to provide a more accurate perturbation estimate and an efficient gradient descent, thereby improving parameter robustness against noise. Notably, DPAdapter can act as a plug-and-play component and be combined with existing DPML algorithms to further improve their performance. Our experiments show that DPAdapter vastly enhances state-of-the-art DPML algorithms, increasing average accuracy from 72.92\% to 77.09\% with a privacy budget of $ε=4$.

cs.LG↗
arXiv2024

Gradient Shaping: Enhancing Backdoor Attack Against Reverse Engineering

Most existing methods to detect backdoored machine learning (ML) models take one of the two approaches: trigger inversion (aka. reverse engineer) and weight analysis (aka. model diagnosis). In particular, the gradient-based trigger inversion is considered to be among the most effective backdoor detection techniques, as evidenced by the TrojAI competition, Trojan Detection Challenge and backdoorBench. However, little has been done to understand why this technique works so well and, more importantly, whether it raises the bar to the backdoor attack. In this paper, we report the first attempt to answer this question by analyzing the change rate of the backdoored model around its trigger-carrying inputs. Our study shows that existing attacks tend to inject the backdoor characterized by a low change rate around trigger-carrying inputs, which are easy to capture by gradient-based trigger inversion. In the meantime, we found that the low change rate is not necessary for a backdoor attack to succeed: we design a new attack enhancement called \textit{Gradient Shaping} (GRASP), which follows the opposite direction of adversarial training to reduce the change rate of a backdoored model with regard to the trigger, without undermining its backdoor effect. Also, we provide a theoretic analysis to explain the effectiveness of this new technique and the fundamental weakness of gradient-based trigger inversion. Finally, we perform both theoretical and experimental analysis, showing that the GRASP enhancement does not reduce the effectiveness of the stealthy attacks against the backdoor detection methods based on weight analysis, as well as other backdoor mitigation methods without using detection.

cs.CR↗
arXiv2024

Smart patterning for topological pumping of elastic surface waves

Topological pumping supplies a robust mechanism to steer waves across a sample without being affected by disorders and defects. For the first time, we demonstrate the pumping of elastic surface waves, achieved by a smart patterning of a surface that creates a synthetic dimension, which is explored by the wave as it is launched perpendicularly to the steering direction. Specifically, we design and fabricate an elastic medium decorated with arrays of pillar-type resonators whose eigenmodes are locate below the sound cone, together with coupling bridges edged according to a specific algorithm. We establish a connection between the collective dynamics of the pillars and that of electrons in a magnetic field by deriving an accurate tight-binding model and developing a WKB-type analysis suitable for such discrete aperiodic systems with spatially slow-varying couplings. This enable us to predict topological pumping pattern, which is numerically and experimentally demonstrated by steering waves from one edge of the system to the other. Finally, the immune character of the topologically pumped surface waves against disorder and defects is evidenced. The principle of surface patterning together with the WKB-analysis could provide a powerful new platform for surface wave control and exploration of topological matter in higher dimensions.

physics.app-ph↗
arXiv2024

CatNorth: An Improved Gaia DR3 Quasar Candidate Catalog with Pan-STARRS1 and CatWISE

A complete and pure sample of quasars with accurate redshifts is crucial for quasar studies and cosmology. In this paper, we present CatNorth, an improved Gaia DR3 quasar candidate catalog with more than 1.5 million sources in the 3$π$ sky built with data from Gaia, Pan-STARRS1, and CatWISE2020. The XGBoost algorithm is used to reclassify the original Gaia DR3 quasar candidates as stars, galaxies, and quasars. To construct training/validation datasets for the classification, we carefully built two different master stellar samples in addition to the spectroscopic galaxy and quasar samples. An ensemble classification model is obtained by averaging two XGBoost classifiers trained with different master stellar samples. Using a probability threshold of $p_{\mathrm{QSO\_mean}}>0.95$ in our ensemble classification model and an additional cut on the logarithmic probability density of zero proper motion, we retrieved 1,545,514 reliable quasar candidates from the parent Gaia DR3 quasar candidate catalog. We provide photometric redshifts for all candidates with an ensemble regression model. For a subset of 89,100 candidates, accurate spectroscopic redshifts are estimated with the Convolutional Neural Network from the Gaia BP/RP spectra. The CatNorth catalog has a high purity of ~ 90% while maintaining high completeness, which is an ideal sample to understand the quasar population and its statistical properties. The CatNorth catalog is used as the main source of input catalog for the LAMOST phase III quasar survey, which is expected to build a highly complete sample of bright quasars with $i < 19.5$.

astro-ph.GA↗
arXiv2024

Non-orthogonal cavity modes near exceptional points in the far field

Non-orthogonal eigenstates are a fundamental feature of non-Hermitian systems and are accompanied by the emergence of nontrivial features. However, the platforms to explore non-Hermitian mode couplings mainly measure near-field effects, and the far-field behaviour remain mostly unexplored. Here, we study how a microcavity with non-Hermitian mode coupling exhibits eigenstate non-orthogonality by investigating the spatial field and the far-field polarization of cavity modes. The non-Hermiticity arises from asymmetric backscattering, which is controlled by integrating two scatterers of different size and location into a microdisk. We observe that the spatial field overlaps of two modes increases abruptly to its maximum value, whilst different far-field elliptical polarizations of two modes coalesce when approaching an exceptional point. We demonstrate such features experimentally by measuring the far-field polarization from the fabricated microdisks. Our work reveals the non-orthogonality in the far-field degree of freedom, and the integrability of the microdisks paves a way to integrate more non-Hermitian optical properties into nanophotonic systems.

physics.optics↗
arXiv2023

Can Large Language Models Be Good Companions? An LLM-Based Eyewear System with Conversational Common Ground

Developing chatbots as personal companions has long been a goal of artificial intelligence researchers. Recent advances in Large Language Models (LLMs) have delivered a practical solution for endowing chatbots with anthropomorphic language capabilities. However, it takes more than LLMs to enable chatbots that can act as companions. Humans use their understanding of individual personalities to drive conversations. Chatbots also require this capability to enable human-like companionship. They should act based on personalized, real-time, and time-evolving knowledge of their owner. We define such essential knowledge as the \textit{common ground} between chatbots and their owners, and we propose to build a common-ground-aware dialogue system from an LLM-based module, named \textit{OS-1}, to enable chatbot companionship. Hosted by eyewear, OS-1 can sense the visual and audio signals the user receives and extract real-time contextual semantics. Those semantics are categorized and recorded to formulate historical contexts from which the user's profile is distilled and evolves over time, i.e., OS-1 gradually learns about its user. OS-1 combines knowledge from real-time semantics, historical contexts, and user-specific profiles to produce a common-ground-aware prompt input into the LLM module. The LLM's output is converted to audio, spoken to the wearer when appropriate.We conduct laboratory and in-field studies to assess OS-1's ability to build common ground between the chatbot and its user. The technical feasibility and capabilities of the system are also evaluated. OS-1, with its common-ground awareness, can significantly improve user satisfaction and potentially lead to downstream tasks such as personal emotional support and assistance.

cs.HC↗