arXiv subjects

Fan Wu

Publications and source records attributed to Fan Wu.

At least 307 records · Page 17Linked to original sources

arXiv2022

Malware Detection and Prevention using Artificial Intelligence Techniques

With the rapid technological advancement, security has become a major issue due to the increase in malware activity that poses a serious threat to the security and safety of both computer systems and stakeholders. To maintain stakeholders, particularly, end users security, protecting the data from fraudulent efforts is one of the most pressing concerns. A set of malicious programming code, scripts, active content, or intrusive software that is designed to destroy intended computer systems and programs or mobile and web applications is referred to as malware. According to a study, naive users are unable to distinguish between malicious and benign applications. Thus, computer systems and mobile applications should be designed to detect malicious activities towards protecting the stakeholders. A number of algorithms are available to detect malware activities by utilizing novel concepts including Artificial Intelligence, Machine Learning, and Deep Learning. In this study, we emphasize Artificial Intelligence (AI) based techniques for detecting and preventing malware activity. We present a detailed review of current malware detection technologies, their shortcomings, and ways to improve efficiency. Our study shows that adopting futuristic approaches for the development of malware detection applications shall provide significant advantages. The comprehension of this synthesis shall help researchers for further research on malware detection and prevention using AI.

cs.CR↗

arXiv2022

Ambiguity Tube MPC

This paper is about a class of distributionally robust model predictive controllers (MPC) for nonlinear stochastic processes that evaluate risk and control performance measures by propagating ambiguity sets in the space of state probability measures. A framework for formulating such ambiguity tube MPC controllers is presented, which is based on modern measure-theoretic methods from the field of optimal transport theory. Moreover, a supermartingale based analysis technique is proposed, leading to stochastic stability results for a large class of distributionally robust controllers for linear and nonlinear systems. In this context, we also discuss how to construct terminal cost functions for stochastic and distributionally robust MPC that ensure closed-loop stability and asymptotic convergence to robust invariant sets. The corresponding theoretical developments are illustrated by tutorial-style examples and a numerical case study.

math.OC↗

arXiv2022

Walle: An End-to-End, General-Purpose, and Large-Scale Production System for Device-Cloud Collaborative Machine Learning

To break the bottlenecks of mainstream cloud-based machine learning (ML) paradigm, we adopt device-cloud collaborative ML and build the first end-to-end and general-purpose system, called Walle, as the foundation. Walle consists of a deployment platform, distributing ML tasks to billion-scale devices in time; a data pipeline, efficiently preparing task input; and a compute container, providing a cross-platform and high-performance execution environment, while facilitating daily task iteration. Specifically, the compute container is based on Mobile Neural Network (MNN), a tensor compute engine along with the data processing and model execution libraries, which are exposed through a refined Python thread-level virtual machine (VM) to support diverse ML tasks and concurrent task execution. The core of MNN is the novel mechanisms of operator decomposition and semi-auto search, sharply reducing the workload in manually optimizing hundreds of operators for tens of hardware backends and further quickly identifying the best backend with runtime optimization for a computation graph. The data pipeline introduces an on-device stream processing framework to enable processing user behavior data at source. The deployment platform releases ML tasks with an efficient push-then-pull method and supports multi-granularity deployment policies. We evaluate Walle in practical e-commerce application scenarios to demonstrate its effectiveness, efficiency, and scalability. Extensive micro-benchmarks also highlight the superior performance of MNN and the Python thread-level VM. Walle has been in large-scale production use in Alibaba, while MNN has been open source with a broad impact in the community.

cs.LG↗

arXiv2022

Energy conservation for the non-resistive MHD equations with physical boundaries

In this paper, we study the energy equality for weak solutions to the non-resistive MHD equations with physical boundaries. Although the equations of magnetic field $b$ are of hyperbolic type, and the boundary effects are considered, we still prove the global energy equality provided that $ u \in L^{q}_{loc}\left(0, T ; L^{p}(Ω)\right) \text { for any } \frac{1}{q}+\frac{1}{p} \leq \frac{1}{2}, \text { with } p \geq 4,\text{ and } b \in L^{r}_{loc}\left(0, T ; L^{s}(Ω)\right) \text { for any } \frac{1}{r}+\frac{1}{s} \leq \frac{1}{2}, \text { with } s \geq 4 $. In particular, compared with the existed results, we do not require any boundary layer assumptions and additional conditions on the pressure $P$. Our result requires the regularity of boundary $\partialΩ$ is only Lipschitz which is the minimum requirement to make the boundary condition $b\cdot n$ sense. The proof is based on the important properties of weak solutions of the nonstationary Stokes system and the separate mollification of weak solutions from the boundary effect by considering a non-standard local energy equality and transform the boundary effects into the estimates of the gradient of cut-off functions.

math.AP↗

arXiv2022

On the energy equality for the 3D incompressible viscoelastic flows

In this paper, we study the problem of energy conservation for the solutions to the incompressible viscoelastic flows. First, we consider Leray-Hopf weak solutions in the bounded Lipschitz domain $Ω$ in $\mathbb{R}^d\,\, (d\geq 2)$. We prove that under the Shinbrot type conditions $ u \in L^{q}_{loc}\left(0, T ; L^{p}(Ω)\right) \text { for any } \frac{1}{q}+\frac{1}{p} \leq \frac{1}{2}, \text { with } p \geq 4,\text{ and } {\bf F} \in L^{r}_{loc}\left(0, T ; L^{s}(Ω)\right) \text { for any } \frac{1}{r}+\frac{1}{s} \leq \frac{1}{2}, \text { with } s \geq 4 $, the boundary conditions $u|_{\partialΩ}=0,\,\,{\bf F}\cdot n|_{\partialΩ}=0$ can inhibit the boundary effect and guarantee the validity of energy equality. Next, we apply this idea to deal with the case $Ω= \mathbb{R}^d\,\,(d=2, 3, 4)$, and showed that the energy is conserved for $u\in L_{loc}^{q}\left(0,T;L_{loc}^{p}\left(\mathbb{R}^{d}\right)\right)$ with $ \frac{2}{q}+\frac{2}{p}\leq1, p\geq 4 $ and $ {\bf F}\in L_{loc}^{r}\left(0,T;L_{loc}^{s}\left(\mathbb{R}^{d}\right)\right)\cap L^{\frac{4d+8}{d+4}}\left(0,T;L^{\frac{4d+8}{d+4}}\left(\mathbb{R}^{d}\right)\right)$ with $\frac{2}{r}+\frac{2}{s}\leq1, s\geq 4 $. This result shows that the behavior of solutions in the finite regions and the behavior at infinite play different roles in the energy conservation. Finally, we consider the problem of energy conservation for distributional solutions and show energy equality for the distributional solutions belonging to the so-called Lions class $L^4L^4$.

math.AP↗

arXiv2022

Clone Detection on Large Scala Codebases

Code clones are identical or similar code segments. The wide existence of code clones can increase the cost of maintenance and jeopardise the quality of software. The research community has developed many techniques to detect code clones, however, there is little evidence of how these techniques may perform in industrial use cases. In this paper, we aim to uncover the differences when such techniques are applied in industrial use cases. We conducted large scale experimental research on the performance of two state-of-the-art code clone detection techniques, SourcererCC and AutoenCODE, on both open source projects and an industrial project written in the Scala language. Our results reveal that both algorithms perform differently on the industrial project, with the largest drop in precision being 30.7\%, and the largest increase in recall being 32.4\%. By manually labelling samples of the industrial project by its developers, we discovered that there are substantially less Type-3 clones in the aforementioned project than that in the open source projects.

cs.SE↗

arXiv2022

DataLens: Scalable Privacy Preserving Training via Gradient Compression and Aggregation

Recent success of deep neural networks (DNNs) hinges on the availability of large-scale dataset; however, training on such dataset often poses privacy risks for sensitive training information. In this paper, we aim to explore the power of generative models and gradient sparsity, and propose a scalable privacy-preserving generative model DATALENS. Comparing with the standard PATE privacy-preserving framework which allows teachers to vote on one-dimensional predictions, voting on the high dimensional gradient vectors is challenging in terms of privacy preservation. As dimension reduction techniques are required, we need to navigate a delicate tradeoff space between (1) the improvement of privacy preservation and (2) the slowdown of SGD convergence. To tackle this, we take advantage of communication efficient learning and propose a novel noise compression and aggregation approach TOPAGG by combining top-k compression for dimension reduction with a corresponding noise injection mechanism. We theoretically prove that the DATALENS framework guarantees differential privacy for its generated data, and provide analysis on its convergence. To demonstrate the practical usage of DATALENS, we conduct extensive experiments on diverse datasets including MNIST, Fashion-MNIST, and high dimensional CelebA, and we show that, DATALENS significantly outperforms other baseline DP generative models. In addition, we adapt the proposed TOPAGG approach, which is one of the key building blocks in DATALENS, to DP SGD training, and show that it is able to achieve higher utility than the state-of-the-art DP SGD approach in most cases. Our code is publicly available at https://github.com/AI-secure/DataLens.

cs.LG↗

arXiv2022

CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothing

As reinforcement learning (RL) has achieved great success and been even adopted in safety-critical domains such as autonomous vehicles, a range of empirical studies have been conducted to improve its robustness against adversarial attacks. However, how to certify its robustness with theoretical guarantees still remains challenging. In this paper, we present the first unified framework CROP (Certifying Robust Policies for RL) to provide robustness certification on both action and reward levels. In particular, we propose two robustness certification criteria: robustness of per-state actions and lower bound of cumulative rewards. We then develop a local smoothing algorithm for policies derived from Q-functions to guarantee the robustness of actions taken along the trajectory; we also develop a global smoothing algorithm for certifying the lower bound of a finite-horizon cumulative reward, as well as a novel local smoothing algorithm to perform adaptive search in order to obtain tighter reward certification. Empirically, we apply CROP to evaluate several existing empirically robust RL algorithms, including adversarial training and different robust regularization, in four environments (two representative Atari games, Highway, and CartPole). Furthermore, by evaluating these algorithms against adversarial attacks, we demonstrate that our certification are often tight. All experiment results are available at website https://crop-leaderboard.github.io.

cs.LG↗

arXiv2022

COPA: Certifying Robust Policies for Offline Reinforcement Learning against Poisoning Attacks

As reinforcement learning (RL) has achieved near human-level performance in a variety of tasks, its robustness has raised great attention. While a vast body of research has explored test-time (evasion) attacks in RL and corresponding defenses, its robustness against training-time (poisoning) attacks remains largely unanswered. In this work, we focus on certifying the robustness of offline RL in the presence of poisoning attacks, where a subset of training trajectories could be arbitrarily manipulated. We propose the first certification framework, COPA, to certify the number of poisoning trajectories that can be tolerated regarding different certification criteria. Given the complex structure of RL, we propose two certification criteria: per-state action stability and cumulative reward bound. To further improve the certification, we propose new partition and aggregation protocols to train robust policies. We further prove that some of the proposed certification methods are theoretically tight and some are NP-Complete problems. We leverage COPA to certify three RL environments trained with different algorithms and conclude: (1) The proposed robust aggregation protocols such as temporal aggregation can significantly improve the certifications; (2) Our certification for both per-state action stability and cumulative reward bound are efficient and tight; (3) The certification for different training algorithms and environments are different, implying their intrinsic robustness properties. All experimental results are available at https://copa-leaderboard.github.io.

cs.LG↗

arXiv2022

Quasi 1D electronic transport in a 2D magnetic semiconductor

We investigate electronic transport through exfoliated multilayers of CrSBr, a 2D semiconductor that is attracting attention because of its magnetic properties. We find an extremely pronounced anisotropy that manifests itself in qualitative and quantitative differences of all quantities measured along the in-plane \textit{a} and \textit{b} crystallographic directions. In particular, we observe a qualitatively different dependence of the conductivities $σ_a$ and $σ_b$ on temperature and gate voltage, accompanied by orders of magnitude differences in their values ($σ_b$/$σ_a \approx 3\cdot10^2-10^5$ at low temperature and large negative gate voltage). We also find a different behavior of the longitudinal magnetoresistance in the two directions, and the complete absence of the Hall effect in transverse resistance measurements. These observations appear not to be compatible with a description in terms of conventional band transport of a 2D doped semiconductor. The observed phenomenology -- together with unambiguous signatures of a 1D van Hove singularity that we detect in energy resolved photocurrent measurements -- indicate that electronic transport through CrSBr multilayers is better interpreted by considering the system as formed by weakly and incoherently coupled 1D wires, than by conventional 2D band transport. We conclude that CrSBr is the first 2D semiconductor to show distinctly quasi 1D electronic transport properties.

cond-mat.mes-hall↗

arXiv2022

Direct Observation of Thermalization to a Rayleigh-Jeans Distribution in Multimode Optical Fibers

Recent years have witnessed a resurgence of interest in nonlinear multimode optical systems where a host of intriguing effects have been observed that are impossible in single-mode settings. While nonlinearity can provide a rich environment where the chaotic power exchange among thousands of modes can lead to novel behaviors, at the same time, it poses a major challenge in terms of understanding and harnessing these processes to advantage. Over the years, statistical models have been developed to macroscopically describe the response of these complex systems. One of the cornerstones of these theoretical formalisms is the prediction of a photon-photon mediated thermalization process that leads to a Rayleigh-Jeans distribution of mode occupations. Here we report the use of mode-resolved measurement techniques to make the first direct observations of thermalization to a Rayleigh-Jeans power distribution in a multimode optical fiber. We experimentally demonstrate that the underlying system Hamiltonian remains invariant during propagation while power equipartition takes place among degenerate groups of modes - all in full accord with theoretical predictions. Our results may pave the way toward a new generation of high-power optical sources whose brightness and modal content can be controlled using principles from thermodynamics and statistical mechanics.

physics.optics↗

arXiv2022

Cryptocurrency Trading: A Comprehensive Survey

In recent years, the tendency of the number of financial institutions including cryptocurrencies in their portfolios has accelerated. Cryptocurrencies are the first pure digital assets to be included by asset managers. Although they have some commonalities with more traditional assets, they have their own separate nature and their behaviour as an asset is still in the process of being understood. It is therefore important to summarise existing research papers and results on cryptocurrency trading, including available trading platforms, trading signals, trading strategy research and risk management. This paper provides a comprehensive survey of cryptocurrency trading research, by covering 146 research papers on various aspects of cryptocurrency trading (e.g., cryptocurrency trading systems, bubble and extreme conditions, prediction of volatility and return, crypto-assets portfolio construction and crypto-assets, technical trading and others). This paper also analyses datasets, research trends and distribution among research objects(contents/properties) and technologies, concluding with some promising opportunities that remain open in cryptocurrency trading.

q-fin.TR↗

arXiv2022

On-Device Learning with Cloud-Coordinated Data Augmentation for Extreme Model Personalization in Recommender Systems

Data heterogeneity is an intrinsic property of recommender systems, making models trained over the global data on the cloud, which is the mainstream in industry, non-optimal to each individual user's local data distribution. To deal with data heterogeneity, model personalization with on-device learning is a potential solution. However, on-device training using a user's small size of local samples will incur severe overfitting and undermine the model's generalization ability. In this work, we propose a new device-cloud collaborative learning framework, called CoDA, to break the dilemmas of purely cloud-based learning and on-device learning. The key principle of CoDA is to retrieve similar samples from the cloud's global pool to augment each user's local dataset to train the recommendation model. Specifically, after a coarse-grained sample matching on the cloud, a personalized sample classifier is further trained on each device for a fine-grained sample filtering, which can learn the boundary between the local data distribution and the outside data distribution. We also build an end-to-end pipeline to support the flows of data, model, computation, and control between the cloud and each device. We have deployed CoDA in a recommendation scenario of Mobile Taobao. Online A/B testing results show the remarkable performance improvement of CoDA over both cloud-based learning without model personalization and on-device training without data augmentation. Overhead testing on a real device demonstrates the computation, storage, and communication efficiency of the on-device tasks in CoDA.

cs.LG↗

arXiv2022

Hadronic molecule interpretation of $T^+_{cc}$ and its beauty-partners

Motivated by the latest discovery of a new tetraquark $T_{cc}^+$ with two charm quarks and two light antiquarks by LHCb Collaboration, we investigated the $DD^*$ hadronic molecule interpretation of $T_{cc}^+$. By calculation, the mass and the decay width of this new structure $T_{cc}^+$ can be understood in one-meson exchange potential model. The binding energies for these $DD^*$ hadronic molecules with $J^P=1^+$ are around $1MeV$. Besides, we also studied the possible beauty-partners $T_{bb}(10598)$ of hadronic molecule $T^+_{cc}$, which may be feasible in future LHCb experiments.

hep-ph↗

arXiv2022

A Cooperative-Competitive Multi-Agent Framework for Auto-bidding in Online Advertising

In online advertising, auto-bidding has become an essential tool for advertisers to optimize their preferred ad performance metrics by simply expressing high-level campaign objectives and constraints. Previous works designed auto-bidding tools from the view of single-agent, without modeling the mutual influence between agents. In this paper, we instead consider this problem from a distributed multi-agent perspective, and propose a general $\underline{M}$ulti-$\underline{A}$gent reinforcement learning framework for $\underline{A}$uto-$\underline{B}$idding, namely MAAB, to learn the auto-bidding strategies. First, we investigate the competition and cooperation relation among auto-bidding agents, and propose a temperature-regularized credit assignment to establish a mixed cooperative-competitive paradigm. By carefully making a competition and cooperation trade-off among agents, we can reach an equilibrium state that guarantees not only individual advertiser's utility but also the system performance (i.e., social welfare). Second, to avoid the potential collusion behaviors of bidding low prices underlying the cooperation, we further propose bar agents to set a personalized bidding bar for each agent, and then alleviate the revenue degradation due to the cooperation. Third, to deploy MAAB in the large-scale advertising system with millions of advertisers, we propose a mean-field approach. By grouping advertisers with the same objective as a mean auto-bidding agent, the interactions among the large-scale advertisers are greatly simplified, making it practical to train MAAB efficiently. Extensive experiments on the offline industrial dataset and Alibaba advertising platform demonstrate that our approach outperforms several baseline methods in terms of social welfare and revenue.

cs.MA↗

arXiv2021

Locality Sensitive Hash Aggregated Nonlinear Neighbourhood Matrix Factorization for Online Sparse Big Data Analysis

Matrix factorization (MF) can extract the low-rank features and integrate the information of the data manifold distribution from high-dimensional data, which can consider the nonlinear neighbourhood information. Thus, MF has drawn wide attention for low-rank analysis of sparse big data, e.g., Collaborative Filtering (CF) Recommender Systems, Social Networks, and Quality of Service. However, the following two problems exist: 1) huge computational overhead for the construction of the Graph Similarity Matrix (GSM), and 2) huge memory overhead for the intermediate GSM. Therefore, GSM-based MF, e.g., kernel MF, graph regularized MF, etc., cannot be directly applied to the low-rank analysis of sparse big data on cloud and edge platforms. To solve this intractable problem for sparse big data analysis, we propose Locality Sensitive Hashing (LSH) aggregated MF (LSH-MF), which can solve the following problems: 1) The proposed probabilistic projection strategy of LSH-MF can avoid the construction of the GSM. Furthermore, LSH-MF can satisfy the requirement for the accurate projection of sparse big data. 2) To run LSH-MF for fine-grained parallelization and online learning on GPUs, we also propose CULSH-MF, which works on CUDA parallelization. Experimental results show that CULSH-MF can not only reduce the computational time and memory overhead but also obtain higher accuracy. Compared with deep learning models, CULSH-MF can not only save training time but also achieve the same accuracy performance.

cs.DC↗

arXiv2021

On Designing a Two-stage Auction for Online Advertising

For the scalability of industrial online advertising systems, a two-stage auction architecture is widely used to enable efficient ad allocation on a large set of corpus within a limited response time. The current deployed two-stage ad auction usually retrieves an ad subset by a coarse ad quality metric in a pre-auction stage, and then determines the auction outcome by a refined metric in the subsequent stage. However, this simple and greedy solution suffers from serious performance degradation, as it regards the decision in each stage separately, leading to an improper ad selection metric for the pre-auction stage. In this work, we explicitly investigate the relation between the coarse and refined ad quality metrics, and design a two-stage ad auction by taking the decision interaction between the two stages into account. We decouple the design of the two-stage auction by solving a stochastic subset selection problem in the pre-auction stage and conducting a general second price (GSP) auction in the second stage. We demonstrate that this decouple still preserves the incentive compatibility of the auction mechanism. As the proposed formulation of the pre-auction stage is an NP-hard problem, we propose a scalable approximation solution by defining a new subset selection metric, namely Pre-Auction Score (PAS). Experiment results on both public and industrial dataset demonstrate the significant improvement on social welfare and revenue of the proposed two-stage ad auction, than the intuitive greedy two-stage auction and other baselines.

cs.GT↗

arXiv2021

Secure and Efficient Blockchain based Knowledge Sharing for Intelligent Connected Vehicles

The emergence of Intelligent Connected Vehicles (ICVs) shows great potential for future intelligent traffic systems, enhancing both traffic safety and road efficiency. However, the ICVs relying on data driven perception and driving models face many challenges, including the lack of comprehensive knowledge to deal with complicated driving context. In this paper, we are motivated to investigate cooperative knowledge sharing for ICVs. We propose a secure and efficient directed acyclic graph (DAG) blockchain based knowledge sharing framework, aiming to cater for the micro-transaction based vehicular networks. The framework can realize both local and cross-regional knowledge sharing. Then, the framework is applied to autonomous driving applications, wherein machine learning based models for autonomous driving control can be shared. A lightweight tip selection algorithm (TSA) is proposed for the DAG based knowledge sharing framework to achieve consensus and identity verification for cross-regional vehicles. To enhance model accuracy as well as minimizing bandwidth consumption, an adaptive asynchronous distributed learning (ADL) based scheme is proposed for model uploading and downloading. Experiment results show that the blockchain based knowledge sharing is secure, and it can resist attacks from malicious users. In addition, the proposed adaptive ADL scheme can enhance driving safety related performance compared to several existing algorithms.

cs.NI↗