arXiv subjects

Yu Jiang

Publications and source records attributed to Yu Jiang.

At least 145 records · Page 8Linked to original sources

arXiv2021

A note on modular Terwilliger algebras of association schemes

Let $p$ denote a prime number. In this note, we focus on the modular Terwilliger algebras of association schemes defined in [3]. We define the primary module of a modular Terwilliger algebra of an association scheme and determine all its composition factors up to isomorphism. We then characterize the $p'$-valenced association schemes by some properties of their modular Terwilliger algebras. The corollaries about the modular Terwilliger algebras of association schemes are given.

math.CO↗

arXiv2021

Diffuse optical tomography by simulated annealing via a spin Hamiltonian

Diffuse optical tomography (DOT) is an imaging modality which uses near-infrared light. Although iterative numerical schemes are commonly used for its inverse problem, correct solutions are not obtained unless good initial guesses are chosen. We propose a numerical scheme of DOT which works even when good initial guesses of optical parameters are not available. We use simulated annealing (SA) which is a method of the Markov-chain Monte Carlo. To implement SA for DOT, a spin Hamiltonian is introduced in the cost function, and the Metropolis algorithm or single-component Metropolis-Hastings algorithm is used. By numerical experiments, it is shown that an initial random spin configuration is brought to a converged configuration by SA and targets in the medium are reconstructed. The proposed numerical method solves the inverse problem for DOT by finding the ground state of a spin Hamiltonian with SA.

math.NA↗

arXiv2021

Influence of trapezoidal lobe strut on fuel mixing and combustion in supersonic combustion chamber

The fuel injection system is the main process in supersonic vehicles. In this article, injection of fuel jet through a trapezoidal strut in the coaxial supersonic air stream is investigated. The numerical method is applied for the simulation of three-dimensional trapezoidal lobe strut in supersonic flow. Hydrogen gas is injected from the end of the trapezoidal strut injector to augment fuel mixing and diffusion downstream by intensifying vortices inside the domain. The power of the vortices varies by the injector velocity and free stream condition. To simulate our model, RANS equations with the SST turbulence model are solved. Effects of free-stream velocity and fuel jet total pressure on fuel distribution and mixing are also investigated. Our results indicate that fuel mixing enhances about 18% when a rectangular strut is replaced by a trapezoidal one. An achieved finding disclosed that the strength of vortices is amplified by the presence of a trapezoidal strut injector.

physics.flu-dyn↗

arXiv2021

Spirit Distillation: A Model Compression Method with Multi-domain Knowledge Transfer

Recent applications pose requirements of both cross-domain knowledge transfer and model compression to machine learning models due to insufficient training data and limited computational resources. In this paper, we propose a new knowledge distillation model, named Spirit Distillation (SD), which is a model compression method with multi-domain knowledge transfer. The compact student network mimics out a representation equivalent to the front part of the teacher network, through which the general knowledge can be transferred from the source domain (teacher) to the target domain (student). To further improve the robustness of the student, we extend SD to Enhanced Spirit Distillation (ESD) in exploiting a more comprehensive knowledge by introducing the proximity domain which is similar to the target domain for feature extraction. Results demonstrate that our method can boost mIOU and high-precision accuracy by 1.4% and 8.2% respectively with 78.2% segmentation variance, and can gain a precise compact network with only 41.8% FLOPs.

cs.CV↗

arXiv2021

Spirit Distillation: Precise Real-time Semantic Segmentation of Road Scenes with Insufficient Data

Semantic segmentation of road scenes is one of the key technologies for realizing autonomous driving scene perception, and the effectiveness of deep Convolutional Neural Networks(CNNs) for this task has been demonstrated. State-of-art CNNs for semantic segmentation suffer from excessive computations as well as large-scale training data requirement. Inspired by the ideas of Fine-tuning-based Transfer Learning (FTT) and feature-based knowledge distillation, we propose a new knowledge distillation method for cross-domain knowledge transference and efficient data-insufficient network training, named Spirit Distillation(SD), which allow the student network to mimic the teacher network to extract general features, so that a compact and accurate student network can be trained for real-time semantic segmentation of road scenes. Then, in order to further alleviate the trouble of insufficient data and improve the robustness of the student, an Enhanced Spirit Distillation (ESD) method is proposed, which commits to exploit a more comprehensive general features extraction capability by considering images from both the target and the proximity domains as input. To our knowledge, this paper is a pioneering work on the application of knowledge distillation to few-shot learning. Persuasive experiments conducted on Cityscapes semantic segmentation with the prior knowledge transferred from COCO2017 and KITTI demonstrate that our methods can train a better student network (mIOU and high-precision accuracy boost by 1.4% and 8.2% respectively, with 78.2% segmentation variance) with only 41.8% FLOPs (see Fig. 1).

cs.CV↗

arXiv2021

HDTest: Differential Fuzz Testing of Brain-Inspired Hyperdimensional Computing

Brain-inspired hyperdimensional computing (HDC) is an emerging computational paradigm that mimics brain cognition and leverages hyperdimensional vectors with fully distributed holographic representation and (pseudo)randomness. Compared to other machine learning (ML) methods such as deep neural networks (DNNs), HDC offers several advantages including high energy efficiency, low latency, and one-shot learning, making it a promising alternative candidate on a wide range of applications. However, the reliability and robustness of HDC models have not been explored yet. In this paper, we design, implement, and evaluate HDTest to test HDC model by automatically exposing unexpected or incorrect behaviors under rare inputs. The core idea of HDTest is based on guided differential fuzz testing. Guided by the distance between query hypervector and reference hypervector in HDC, HDTest continuously mutates original inputs to generate new inputs that can trigger incorrect behaviors of HDC model. Compared to traditional ML testing methods, HDTest does not need to manually label the original input. Using handwritten digit classification as an example, we show that HDTest can generate thousands of adversarial inputs with negligible perturbations that can successfully fool HDC models. On average, HDTest can generate around 400 adversarial inputs within one minute running on a commodity computer. Finally, by using the HDTest-generated inputs to retrain HDC models, we can strengthen the robustness of HDC models. To the best of our knowledge, this paper presents the first effort in systematically testing this emerging brain-inspired computational model.

cs.NE↗

arXiv2021

Industry Practice of Coverage-Guided Enterprise-Level DBMS Fuzzing

As an infrastructure for data persistence and analysis, Database Management Systems (DBMSs) are the cornerstones of modern enterprise software. To improve their correctness, the industry has been applying blackbox fuzzing for decades. Recently, the research community achieved impressive fuzzing gains using coverage guidance. However, due to the complexity and distributed nature of enterprise-level DBMSs, seldom are these researches applied to the industry. In this paper, we apply coverage-guided fuzzing to enterprise-level DBMSs from Huawei and Bloomberg LP. In our practice of testing GaussDB and Comdb2, we found major challenges in all three testing stages. The challenges are collecting precise coverage, optimizing fuzzing performance, and analyzing root causes. In search of a general method to overcome these challenges, we propose Ratel, a coverage-guided fuzzer for enterprise-level DBMSs. With its industry-oriented design, Ratel improves the feedback precision, enhances the robustness of input generation, and performs an on-line investigation on the root cause of bugs. As a result, Ratel outperformed other fuzzers in terms of coverage and bugs. Compared to industrial black box fuzzers SQLsmith and SQLancer, as well as coverage-guided academic fuzzer Squirrel, Ratel covered 38.38%, 106.14%, 583.05% more basic blocks than the best results of other three fuzzers in GaussDB, PostgreSQL, and Comdb2, respectively. More importantly, Ratel has discovered 32, 42, and 5 unknown bugs in GaussDB, Comdb2, and PostgreSQL.

cs.SE↗

arXiv2021

IntelliGen: Automatic Driver Synthesis for FuzzTesting

Fuzzing is a technique widely used in vulnerability detection. The process usually involves writing effective fuzz driver programs, which, when done manually, can be extremely labor intensive. Previous attempts at automation leave much to be desired, in either degree of automation or quality of output. In this paper, we propose IntelliGen, a framework that constructs valid fuzz drivers automatically. First, IntelliGen determines a set of entry functions and evaluates their respective chance of exhibiting a vulnerability. Then, IntelliGen generates fuzz drivers for the entry functions through hierarchical parameter replacement and type inference. We implemented IntelliGen and evaluated its effectiveness on real-world programs selected from the Android Open-Source Project, Google's fuzzer-test-suite and industrial collaborators. IntelliGen covered on average 1.08X-2.03X more basic blocks and 1.36X-2.06X more paths over state-of-the-art fuzz driver synthesizers FUDGE and FuzzGen. IntelliGen performed on par with manually written drivers and found 10 more bugs.

cs.SE↗

arXiv2021

V-Gas: Generating High Gas Consumption Inputs to Avoid Out-of-Gas Vulnerability

The out-of-gas error occurs when smart contract programs are provided with inputs that cause excessive gas consumption, and would be easily exploited to make the DoS attack. Multiple approaches have been proposed to estimate the gas limit of a function in smart contracts to avoid such error. However, under estimation often happens when the contract is complicated. In this work, we propose V-Gas, which could automatically generate inputs that maximizes the gas cost and reduce the under estimation cases. V-Gas is designed based on feedback-directed mutational fuzz testing. First, V-Gas builds the gas weighted control flow graph (CFG) of functions in smart contracts. Then, V-Gas develops gas consumption guided selection and mutation strategies to generate the input that maximize the gas consumption. For evaluation, we implement V-Gas based on js-evm, a widely used ethereum virtual machine written in javascript, and conduct experiments on 736 real-world transactions recorded on Ethereum. 44.02\% of the transactions would have out-of-gas errors under the estimation results given by solc, means that the recorded real gas consumption for those recorded transactions is larger than the gas limit value estimated by solc. While V-Gas could reduce the under estimation ratio to 13.86\%. Furthermore, V-Gas has exposed 25 previously unknown out-of-gas vulnerabilities in those widely-used smart contracts, 5 of which have been assigned unique CVE identifiers in the US National Vulnerability Database.

cs.CR↗

arXiv2021

On some trivial source Specht modules

The paper presented here focuses on the classification of trivial source Specht modules. We completely classify the trivial source Specht modules labelled by hook partitions. We also classify the trivial source Specht modules labelled by two-part partitions in the odd characteristic case. Moreover, in the even characteristic case, we prove a result for the classification of the trivial source Specht modules labelled by partitions with 2-weight 2, which justifies a conjecture of [16].

math.RT↗

arXiv2021

On Terwilliger $\mathbb{F}$-algebras of quasi-thin association schemes

In [3], Hanaki defined the Terwilliger algebras of association schemes over a commutative unital ring. In this paper, we call the Terwilliger algebras of association schemes over a field $\mathbb{F}$ the Terwilliger $\mathbb{F}$-algebras of association schemes and study the Terwilliger $\mathbb{F}$-algebras of quasi-thin association schemes. As main results, we determine the $\mathbb{F}$-dimensions, the semisimplicity, the Jacobson radicals, and the algebraic structures of the Terwilliger $\mathbb{F}$-algebras of quasi-thin association schemes. We also get some results with independent interests.

math.CO↗

arXiv2021

RNN-Test: Towards Adversarial Testing for Recurrent Neural Network Systems

While massive efforts have been investigated in adversarial testing of convolutional neural networks (CNN), testing for recurrent neural networks (RNN) is still limited and leaves threats for vast sequential application domains. In this paper, we propose an adversarial testing framework RNN-Test for RNN systems, focusing on the main sequential domains, not only classification tasks. First, we design a novel search methodology customized for RNN models by maximizing the inconsistency of RNN states to produce adversarial inputs. Next, we introduce two state-based coverage metrics according to the distinctive structure of RNNs to explore more inference logics. Finally, RNN-Test solves the joint optimization problem to maximize state inconsistency and state coverage, and crafts adversarial inputs for various tasks of different kinds of inputs. For evaluations, we apply RNN-Test on three sequential models of common RNN structures. On the tested models, the RNN-Test approach is demonstrated to be competitive in generating adversarial inputs, outperforming FGSM-based and DLFuzz-based methods to reduce the model performance more sharply with 2.78% to 32.5% higher success (or generation) rate. RNN-Test could also achieve 52.65% to 66.45% higher adversary rate on MNIST-LSTM model than relevant work testRNN. Compared with the neuron coverage, the proposed state coverage metrics as guidance excel with 4.17% to 97.22% higher success (or generation) rate.

cs.CL↗

arXiv2021

Managing network congestion with a tradable credit scheme: a trip-based MFD approach

This study investigates the efficiency and effectiveness of an area-based tradable credit scheme (TCS) using the trip-based Macroscopic Fundamental Diagram model for the morning commute problem. In the proposed TCS, the regulator distributes initial credits to all travelers and designs a time-varying and trip length specific credit tariff. Credits are traded between travelers and the regulator via a credit market, and the credit price is determined by the demand and supply of credits. The heterogeneity of travelers is considered in terms of desired arrival time, trip length and departure-time choice preferences. The TCS is incorporated into a day-to-day modelling framework to examine the travelers' learning process, the evolution of network, and the properties of the credit market. The existence of an equilibrium solution and the uniqueness of the credit price at the equilibrium state are established analytically. Furthermore, an open-source simulation framework is developed to validate the analytical properties of the proposed TCS and compare it with alternative control strategies in terms of mobility, network performance, and social welfare. Bayesian optimization is then adopted to optimize the credit toll scheme. The numerical results demonstrate that the proposed TCS outperforms the no-control case and matches the performance of the time-of-day pricing strategy, while maintaining revenue-neutral nature.

cs.GT↗

arXiv2020

Bayesian Optimization of Area-based Road Pricing

This study presents a Bayesian Optimization framework for area- and distance-based time-of-day pricing (TODP) for urban networks. The road pricing optimization problem can reach high level of complexity depending on the pricing scheme considered, its associated detailed network properties and the affected heterogeneous demand features. We consider heterogeneous travellers with individual-specific trip attributes and departure-time choice parameters together with a Macroscopic Fundamental Diagram (MFD) model for the urban network. Its mathematical formulation is presented and an agent-based simulation framework is constructed as evaluation function for the TODP optimization problem. The latter becomes highly nonlinear and relying on an expensive-to-evaluate objective function. We then present and test a Bayesian Optimization approach to compute different time-of-day pricing schemes by maximizing social welfare. Our proposed method learns the relationship between the prices and welfare within a few iterations and is able to find good solutions even in scenarios with high dimensionality in the decision variables space, setting a path for complexity reduction in more realistic road pricing optimization problems. Furthermore and as expected, the simulation results show that TODP improves the social welfare against the no-pricing case.

cs.MA↗

arXiv2020

Convergence of a Periodic Orbit Family Close to Asteroids During a Continuation

In this work, we study the continuation of a periodic orbit on a relatively large scale and discover the existence of convergence under certain conditions, which has profound significance in research on asteroids and can provide a total geometric perspective to understanding the evolution of the dynamic characteristics from a global perspective. Based on the polyhedron model, convergence is derived via a series of theoretical analyses and derivations, which shows that a periodic orbit will evolve into a nearly circular orbit with a normal periodic ratio (e.g., 2:1, 3:2, and 4:3) and almost zero torsion under proper circumstances. As an application of the results developed here, three asteroids, (216) Kleopatra, (22) Kalliope and (433) Eros, are studied, and several representative periodic orbit families are detected, with convergence in three different cases: bidirectional, increasing-directional and decreasing-directional continuation. At the same time, four commonalities among these numerical examples are concluded. First, a (pseudo) tangent bifurcation arises at the cuspidal points during the variations in the periodic ratios in a single periodic orbit family. In addition, these cuspidal points in the periodic ratio coincide with the turning points during the variations in the average radius, the maximal torsion and the maximal radius of curvature. Furthermore, the periodic ratio increases (or decreases) with an increase (or decrease) in the Jacobian constant overall. We find the relationship between periodic ratio and Jacobian constant. The results implies that the periodic ratios for a fixed resonant status have an infimum and a supremum. Finally, if the periodic orbit converges to a point, it can be an unstable equilibrium point of the asteroid.

astro-ph.EP↗

arXiv2020

Motion of Dust Ejected from the Surface of Asteroid (101955) Bennu

From Jan. 6, 2019 to Feb. 18, 2019, OSIRIS-REx observed asteroid (101955) Bennu ejecting 11 plumes of dust, of which part is escaping and another part is re-captured by the asteroid. The relative magnitudes of the typical forces acting on the emitted dust are quite different from the environments of the planets and other minor planets in the solar system. Here we show that ejected dust grains from the surface of Bennu can be caught in the gravitational field of Bennu. To this end, we calculated numerically the trajectories of dust grains of various sizes, from the 0.1mum to the ten millimeter range. The shape and the fate of an emitted cloud of particles depend on the size of the grains: smaller grains form a more narrowly confined dust trail while trails formed by larger grains disperse more rapidly. Four different fates are possible for ejected dust. All grains with radius less than 1.0mum, directly re-impact on Bennu or they escape directly. In contrast, a fraction of grains with a radius larger than 10.0 mum will impact or escape only after performing a number of non-Keplerian revolutions around Bennu. Our findings show how dust grains may populate the vicinity of Bennu and other active asteroids and that they can reach interplanetary space and other celestial bodies, implying that organic matter can be transported from carbonaceous asteroids to other celestial bodies, including Earth.

astro-ph.EP↗

arXiv2020

On some $p$-transitive association schemes

In this paper, for any prime $p$, we propose the notion of a $p$-transitive association scheme. This notion aims to generalize the fact that the regular module of a group algebra of a finite group has a unique trivial submodule to the case of the regular modules of modular adjacency algebras. We completely determine the $p$-transitive quasi-thin association schemes and the $p$-transitive association schemes with thin thin residue by their structure theory properties.

math.CO↗

arXiv2020

Debris cloud of India Anti-Satellite Test to Microsat-R Satellite

Understanding the motion of debris cloud produced by the anti-satellite test can help us to know the danger of these tests. This study presents the orbit status of 57 fragments observed by the CelesTrak and presented in the NORAD Two-Line Element Sets of India Anti-Satellite Test. There are 10 of these observed fragments have altitudes of the apogee larger than 1000.0km, the maximum one is 1725.7km. We also numerical calculated the number of debris, the results show that the number of debris with the diameter larger than 0.2m is 14, the number of debris with the diameter larger than 0.01m is 6587, and the number of debris with the diameter larger than 0.001m is 7.22e+5. The results of the secondary collision of the debris will produced more fragments in the space. The life time of the fragments depends on the initial orbit parameters and the sizes of the debris.

astro-ph.EP↗