arXiv subjects

Zhi Wang

Publications and source records attributed to Zhi Wang.

At least 253 records · Page 14Linked to original sources

arXiv2022

DeepC2: AI-powered Covert Command and Control on OSNs

Command and control (C&C) is important in an attack. It transfers commands from the attacker to the malware in the compromised hosts. Currently, some attackers use online social networks (OSNs) in C&C tasks. There are two main problems in the C&C on OSNs. First, the process for the malware to find the attacker is reversible. If the malware sample is analyzed by the defender, the attacker would be exposed before publishing the commands. Second, the commands in plain or encrypted form are regarded as abnormal contents by OSNs, which would raise anomalies and trigger restrictions on the attacker. The defender can limit the attacker once it is exposed. In this work, we propose DeepC2, an AI-powered C&C on OSNs, to solve these problems. For the reversible hard-coding, the malware finds the attacker using a neural network model. The attacker's avatars are converted into a batch of feature vectors, and the defender cannot recover the avatars in advance using the model and the feature vectors. To solve the abnormal contents on OSNs, hash collision and text data augmentation are used to embed commands into normal contents. The experiment on Twitter shows that command-embedded tweets can be generated efficiently. The malware can find the attacker covertly on OSNs. Security analysis shows it is hard to recover the attacker's identifiers in advance.

cs.CR↗

arXiv2022

EvilModel 2.0: Bringing Neural Network Models into Malware Attacks

Security issues have gradually emerged with the continuous development of artificial intelligence (AI). Earlier work verified the possibility of converting neural network models into stegomalware, embedding malware into a model with limited impact on the model's performance. However, existing methods are not applicable in real-world attack scenarios and do not attract enough attention from the security community due to performance degradation and additional workload. Therefore, we propose an improved stegomalware EvilModel. By analyzing the composition of the neural network model, three new methods for embedding malware into the model are proposed: MSB reservation, fast substitution, and half substitution, which can embed malware that accounts for half of the model's volume without affecting the model's performance. We built 550 EvilModels using ten mainstream neural network models and 19 malware samples. The experiment shows that EvilModel achieved an embedding rate of 48.52\%. A quantitative algorithm is proposed to evaluate the existing embedding methods. We also design a trigger and propose a threat scenario for the targeted attack. The practicality and effectiveness of the proposed methods were demonstrated by experiments and analyses of the embedding capacity, performance impact, and detection evasion.

cs.CR↗

arXiv2022

Density Functional description of spin, lattice, and spin-lattice dynamics in antiferromagnetic and paramagnetic phases at finite temperatures

Describing the (a) electronic and magnetic properties (EMP) of antiferromagnetic or paramagnetic phases of compounds generally requires the knowledge of (b) the spin configurations and lattice structure (SCLS) of such phases at a given temperature. Indeed, studying the coupling between (a) and (b) has been an outstanding challenge in the theory of magnetism. The traditional approach to electronic phases of matter has generally focused on solving the problem of EMP regarding the SCLS as a spectator degree of freedom (DOF). Yet, it has been recognized that EMP of a compound generally respond self-consistently to changes in SCLS and vice versa. We construct here a practical, density functional theory (DFT)-based approach that provides the SCLS as a function of temperature, involving the description of spin, lattice, and spin-lattice dynamics of different magnetic phases. We distinguish three Levels of dynamics: (I) dynamics of the spin DOF treated via noncollinear Heisenberg Monte-Carlo with exchange energies from DFT, (II) dynamics of the lattice DOF treated by ab initio molecular dynamics (AIMD) employing a fixed spin configuration from Level I at the simulated temperature, and (III) coupling of spin and lattice dynamics via Landau-Lifshitz-Gilbert spin dynamics combined with AIMD. Such SCLS at each of the three levels are used as inputs to DFT supercell calculations, providing the EMP at each temperature. The results of this sequence include electronic band structures, band gaps, density of states, as well as the statistical distribution of local moments and the short-range order parameters, each as a function of temperature. Using NiO as a test case, we address the separability of the DOF in magnetic insulators for a minimal description of electronic and magnetic properties, demonstrating that inclusion of spin dynamics and, to some level, lattice dynamics is enough to explain the EMP.

physics.comp-ph↗

arXiv2022

Higgs-Leggett mechanism for the elusive 6e superconductivity observed in Kagome vanadium-based superconductors

A recent Little-Parks experiment on Kagome-structured superconductor CsV_3Sb_5 demonstrated remarkable resistance oscillations with period \phi0/3=hc/6e. Here, we perform analysis based on a theory involving three 2e superconductivity (SC) order parameters associated with the three reciprocal lattice vectors which connect M points of the hexagonal Brillouin zone. In a ring geometry we unveil that, as a series of metastable states, phase of one SC order parameter winds 2πmore or less than the other two ones around the ring, which yields local free energy minima at integer multiples of \phi0/3. Intriguingly, the ground-state degeneracy associated with a Z_2 chirality is crucial, and the Higgs-Leggett mechanism stabilizes domain walls (DW) between chiral domains. At low temperatures DW are expelled from the system resulting in free energy minima only at integer multiples of ϕ_0. Our theory explains successfully the 6e SC observed in experiments, which opens a door for approaching rich physics of Kagome superconductors.

cond-mat.supr-con↗

arXiv2022

Thompson Sampling for Robust Transfer in Multi-Task Bandits

We study the problem of online multi-task learning where the tasks are performed within similar but not necessarily identical multi-armed bandit environments. In particular, we study how a learner can improve its overall performance across multiple related tasks through robust transfer of knowledge. While an upper confidence bound (UCB)-based algorithm has recently been shown to achieve nearly-optimal performance guarantees in a setting where all tasks are solved concurrently, it remains unclear whether Thompson sampling (TS) algorithms, which have superior empirical performance in general, share similar theoretical properties. In this work, we present a TS-type algorithm for a more general online multi-task learning protocol, which extends the concurrent setting. We provide its frequentist analysis and prove that it is also nearly-optimal using a novel concentration inequality for multi-task data aggregation at random stopping times. Finally, we evaluate the algorithm on synthetic data and show that the TS-type algorithm enjoys superior empirical performance in comparison with the UCB-based algorithm and a baseline algorithm that performs TS for each individual task without transfer.

cs.LG↗

arXiv2022

HEBO Pushing The Limits of Sample-Efficient Hyperparameter Optimisation

In this work we rigorously analyse assumptions inherent to black-box optimisation hyper-parameter tuning tasks. Our results on the Bayesmark benchmark indicate that heteroscedasticity and non-stationarity pose significant challenges for black-box optimisers. Based on these findings, we propose a Heteroscedastic and Evolutionary Bayesian Optimisation solver (HEBO). HEBO performs non-linear input and output warping, admits exact marginal log-likelihood optimisation and is robust to the values of learned parameters. We demonstrate HEBO's empirical efficacy on the NeurIPS 2020 Black-Box Optimisation challenge, where HEBO placed first. Upon further analysis, we observe that HEBO significantly outperforms existing black-box optimisers on 108 machine learning hyperparameter tuning tasks comprising the Bayesmark benchmark. Our findings indicate that the majority of hyper-parameter tuning tasks exhibit heteroscedasticity and non-stationarity, multi-objective acquisition ensembles with Pareto front solutions improve queried configurations, and robust acquisition maximisers afford empirical advantages relative to their non-robust counterparts. We hope these findings may serve as guiding principles for practitioners of Bayesian optimisation. All code is made available at https://github.com/huawei-noah/HEBO.

cs.LG↗

arXiv2022

CDFKD-MFS: Collaborative Data-free Knowledge Distillation via Multi-level Feature Sharing

Recently, the compression and deployment of powerful deep neural networks (DNNs) on resource-limited edge devices to provide intelligent services have become attractive tasks. Although knowledge distillation (KD) is a feasible solution for compression, its requirement on the original dataset raises privacy concerns. In addition, it is common to integrate multiple pretrained models to achieve satisfactory performance. How to compress multiple models into a tiny model is challenging, especially when the original data are unavailable. To tackle this challenge, we propose a framework termed collaborative data-free knowledge distillation via multi-level feature sharing (CDFKD-MFS), which consists of a multi-header student module, an asymmetric adversarial data-free KD module, and an attention-based aggregation module. In this framework, the student model equipped with a multi-level feature-sharing structure learns from multiple teacher models and is trained together with a generator in an asymmetric adversarial manner. When some real samples are available, the attention module adaptively aggregates predictions of the student headers, which can further improve performance. We conduct extensive experiments on three popular computer visual datasets. In particular, compared with the most competitive alternative, the accuracy of the proposed framework is 1.18\% higher on the CIFAR-100 dataset, 1.67\% higher on the Caltech-101 dataset, and 2.99\% higher on the mini-ImageNet dataset.

cs.CV↗

arXiv2022

A Dirichlet Process Mixture of Robust Task Models for Scalable Lifelong Reinforcement Learning

While reinforcement learning (RL) algorithms are achieving state-of-the-art performance in various challenging tasks, they can easily encounter catastrophic forgetting or interference when faced with lifelong streaming information. In the paper, we propose a scalable lifelong RL method that dynamically expands the network capacity to accommodate new knowledge while preventing past memories from being perturbed. We use a Dirichlet process mixture to model the non-stationary task distribution, which captures task relatedness by estimating the likelihood of task-to-cluster assignments and clusters the task models in a latent space. We formulate the prior distribution of the mixture as a Chinese restaurant process (CRP) that instantiates new mixture components as needed. The update and expansion of the mixture are governed by the Bayesian non-parametric framework with an expectation maximization (EM) procedure, which dynamically adapts the model complexity without explicit task boundaries or heuristics. Moreover, we use the domain randomization technique to train robust prior parameters for the initialization of each task model in the mixture, thus the resulting model can better generalize and adapt to unseen tasks. With extensive experiments conducted on robot navigation and locomotion domains, we show that our method successfully facilitates scalable lifelong RL and outperforms relevant existing methods.

cs.LG↗

arXiv2022

Josephson radiation patterns in underdamped topological Josephson junctions

Josephson radiation is a useful signature for detecting Majorana zero modes in topological superconductors. We study the Josephson radiation of the underdamped topological Josephson junction within the quantum resistively and capacitively shunted junction model. We show that the quantum dynamics of the Majorana two-level system induce oscillatory patterns in the Josephson emission spectra. With the Floquet method, we obtain analytical results for these oscillatory patterns and find that they are well described by Bessel functions. We perform numerical simulations to verify the analytical results and demonstrate that these Bessel radiation patterns exist for a wide range of junction parameters.

cond-mat.supr-con↗

arXiv2022

SeqNet: An Efficient Neural Network for Automatic Malware Detection

Malware continues to evolve rapidly, and more than 450,000 new samples are captured every day, which makes manual malware analysis impractical. However, existing deep learning detection models need manual feature engineering or require high computational overhead for long training processes, which might be laborious to select feature space and difficult to retrain for mitigating model aging. Therefore, a crucial requirement for a detector is to realize automatic and efficient detection. In this paper, we propose a lightweight malware detection model called SeqNet which could be trained at high speed with low memory required on the raw binaries. By avoiding contextual confusion and reducing semantic loss, SeqNet maintains the detection accuracy when reducing the number of parameters to only 136K. We demonstrate the effectiveness of our methods and the low training cost requirement of SeqNet in our experiments. Besides, we make our datasets and codes public to stimulate further academic research.

cs.CR↗

arXiv2022

Cost Effective MLaaS Federation: A Combinatorial Reinforcement Learning Approach

With the advancement of deep learning techniques, major cloud providers and niche machine learning service providers start to offer their cloud-based machine learning tools, also known as machine learning as a service (MLaaS), to the public. According to our measurement, for the same task, these MLaaSes from different providers have varying performance due to the proprietary datasets, models, etc. Federating different MLaaSes together allows us to improve the analytic performance further. However, naively aggregating results from different MLaaSes not only incurs significant momentary cost but also may lead to sub-optimal performance gain due to the introduction of possible false-positive results. In this paper, we propose Armol, a framework to federate the right selection of MLaaS providers to achieve the best possible analytic performance. We first design a word grouping algorithm to unify the output labels across different providers. We then present a deep combinatorial reinforcement learning based-approach to maximize the accuracy while minimizing the cost. The predictions from the selected providers are then aggregated together using carefully chosen ensemble strategies. The real-world trace-driven evaluation further demonstrates that Armol is able to achieve the same accuracy results with $67\%$ less inference cost.

cs.LG↗

arXiv2022

Deepfake Forensics via An Adversarial Game

With the progress in AI-based facial forgery (i.e., deepfake), people are increasingly concerned about its abuse. Albeit effort has been made for training classification (also known as deepfake detection) models to recognize such forgeries, existing models suffer from poor generalization to unseen forgery technologies and high sensitivity to changes in image/video quality. In this paper, we advocate adversarial training for improving the generalization ability to both unseen facial forgeries and unseen image/video qualities. We believe training with samples that are adversarially crafted to attack the classification models improves the generalization ability considerably. Considering that AI-based face manipulation often leads to high-frequency artifacts that can be easily spotted by models yet difficult to generalize, we further propose a new adversarial training method that attempts to blur out these specific artifacts, by introducing pixel-wise Gaussian blurring models. With adversarial training, the classification models are forced to learn more discriminative and generalizable features, and the effectiveness of our method can be verified by plenty of empirical evidence. Our code will be made publicly available.

cs.CV↗

arXiv2022

Arbitrary Bit-width Network: A Joint Layer-Wise Quantization and Adaptive Inference Approach

Conventional model quantization methods use a fixed quantization scheme to different data samples, which ignores the inherent "recognition difficulty" differences between various samples. We propose to feed different data samples with varying quantization schemes to achieve a data-dependent dynamic inference, at a fine-grained layer level. However, enabling this adaptive inference with changeable layer-wise quantization schemes is challenging because the combination of bit-widths and layers is growing exponentially, making it extremely difficult to train a single model in such a vast searching space and use it in practice. To solve this problem, we present the Arbitrary Bit-width Network (ABN), where the bit-widths of a single deep network can change at runtime for different data samples, with a layer-wise granularity. Specifically, first we build a weight-shared layer-wise quantizable "super-network" in which each layer can be allocated with multiple bit-widths and thus quantized differently on demand. The super-network provides a considerably large number of combinations of bit-widths and layers, each of which can be used during inference without retraining or storing myriad models. Second, based on the well-trained super-network, each layer's runtime bit-width selection decision is modeled as a Markov Decision Process (MDP) and solved by an adaptive inference strategy accordingly. Experiments show that the super-network can be built without accuracy degradation, and the bit-widths allocation of each layer can be adjusted to deal with various inputs on the fly. On ImageNet classification, we achieve 1.1% top1 accuracy improvement while saving 36.2% BitOps.

cs.CV↗

arXiv2022

Transport theory for topological Josephson junctions with a Majorana qubit

We construct a semiclassical theory for the transport of topological junctions starting from a microscopic Hamiltonian that comprehensively includes the interplay among the Majorana qubit, the Josephson phase, and the dissipation process. With the path integral approach, we derive a set of semiclassical equations of motion that can be used to calculate the time evolution of the Josephson phase and the Majorana qubit. In the equations we reveal rich dynamical phenomena such as the qubit induced charge pumping, the effective spin-orbit torque, and the Gilbert damping. We demonstrate the influence of these dynamical phenomena on the transport signatures of the junction. We apply the theory to study the Shapiro steps of the junction, and find the suppression of the first Shapiro step due to the dynamical feedback of the Majorana qubit.

cond-mat.supr-con↗

arXiv2022

Instance Weighted Incremental Evolution Strategies for Reinforcement Learning in Dynamic Environments

Evolution strategies (ES), as a family of black-box optimization algorithms, recently emerge as a scalable alternative to reinforcement learning (RL) approaches such as Q-learning or policy gradient, and are much faster when many central processing units (CPUs) are available due to better parallelization. In this paper, we propose a systematic incremental learning method for ES in dynamic environments. The goal is to adjust previously learned policy to a new one incrementally whenever the environment changes. We incorporate an instance weighting mechanism with ES to facilitate its learning adaptation, while retaining scalability of ES. During parameter updating, higher weights are assigned to instances that contain more new knowledge, thus encouraging the search distribution to move towards new promising areas of parameter space. We propose two easy-to-implement metrics to calculate the weights: instance novelty and instance quality. Instance novelty measures an instance's difference from the previous optimum in the original environment, while instance quality corresponds to how well an instance performs in the new environment. The resulting algorithm, Instance Weighted Incremental Evolution Strategies (IW-IES), is verified to achieve significantly improved performance on challenging RL tasks ranging from robot navigation to locomotion. This paper thus introduces a family of scalable ES algorithms for RL domains that enables rapid learning adaptation to dynamic environments.

cs.LG↗

arXiv2022

Online Continual Adaptation with Active Self-Training

Models trained with offline data often suffer from continual distribution shifts and expensive labeling in changing environments. This calls for a new online learning paradigm where the learner can continually adapt to changing environments with limited labels. In this paper, we propose a new online setting -- Online Active Continual Adaptation, where the learner aims to continually adapt to changing distributions using both unlabeled samples and active queries of limited labels. To this end, we propose Online Self-Adaptive Mirror Descent (OSAMD), which adopts an online teacher-student structure to enable online self-training from unlabeled data, and a margin-based criterion that decides whether to query the labels to track changing distributions. Theoretically, we show that, in the separable case, OSAMD has an $O({T}^{2/3})$ dynamic regret bound under mild assumptions, which is aligned with the $Ω(T^{2/3})$ lower bound of online learning algorithms with full labels. In the general case, we show a regret bound of $O({T}^{2/3} + α^* T)$, where $α^*$ denotes the separability of domains and is usually small. Our theoretical results show that OSAMD can fast adapt to changing environments with active queries. Empirically, we demonstrate that OSAMD achieves favorable regrets under changing environments with limited labels on both simulated and real-world data, which corroborates our theoretical findings.

cs.LG↗

arXiv2022

A Systematic Study of Android Non-SDK (Hidden) Service API Security

Android allows apps to communicate with its system services via system service helpers so that these apps can use various functions provided by the system services. Meanwhile, the system services rely on their service helpers to enforce security checks for protection. Unfortunately, the security checks in the service helpers may be bypassed via directly exploiting the non-SDK (hidden) APIs, degrading the stability and posing severe security threats such as privilege escalation, automatic function execution without users' interactions, crashes, and DoS attacks. Google has proposed various approaches to address this problem, e.g., case-by-case fixing the bugs or even proposing a blacklist to block all the non-SDK APIs. However, the developers can still figure out new ways of exploiting these hidden APIs to evade the non-SDKs restrictions. In this paper, we systematically study the vulnerabilities due to the hidden API exploitation and analyze the effectiveness of Google's countermeasures. We aim to answer if there are still vulnerable hidden APIs that can be exploited in the newest Android 12. We develop a static analysis tool called ServiceAudit to automatically mine the inconsistent security enforcement between service helper classes and the hidden service APIs. We apply ServiceAudit to Android 6~12. Our tool discovers 112 vulnerabilities in Android 6 with higher precision than existing approaches. Moreover, in Android 11 and 12, we identify more than 25 hidden APIs with inconsistent protections; however, only one of the vulnerable APIs can lead to severe security problems in Android 11, and none of them work on Android 12.

cs.CR↗

arXiv2022

A Closer Look at Debiased Temporal Sentence Grounding in Videos: Dataset, Metric, and Approach

Temporal Sentence Grounding in Videos (TSGV), which aims to ground a natural language sentence in an untrimmed video, has drawn widespread attention over the past few years. However, recent studies have found that current benchmark datasets may have obvious moment annotation biases, enabling several simple baselines even without training to achieve SOTA performance. In this paper, we take a closer look at existing evaluation protocols, and find both the prevailing dataset and evaluation metrics are the devils that lead to untrustworthy benchmarking. Therefore, we propose to re-organize the two widely-used datasets, making the ground-truth moment distributions different in the training and test splits, i.e., out-of-distribution (OOD) test. Meanwhile, we introduce a new evaluation metric "dR@n,IoU@m" that discounts the basic recall scores to alleviate the inflating evaluation caused by biased datasets. New benchmarking results indicate that our proposed evaluation protocols can better monitor the research progress. Furthermore, we propose a novel causality-based Multi-branch Deconfounding Debiasing (MDD) framework for unbiased moment prediction. Specifically, we design a multi-branch deconfounder to eliminate the effects caused by multiple confounders with causal intervention. In order to help the model better align the semantics between sentence queries and video moments, we enhance the representations during feature encoding. Specifically, for textual information, the query is parsed into several verb-centered phrases to obtain a more fine-grained textual feature. For visual information, the positional information has been decomposed from moment features to enhance representations of moments with diverse locations. Extensive experiments demonstrate that our proposed approach can achieve competitive results among existing SOTA approaches and outperform the base model with great gains.

cs.CV↗