arXiv subjects

Yi Qi

Publications and source records attributed to Yi Qi.

At least 37 records · Page 2Linked to original sources

arXiv2022

A Hierarchical HAZOP-Like Safety Analysis for Learning-Enabled Systems

Hazard and Operability Analysis (HAZOP) is a powerful safety analysis technique with a long history in industrial process control domain. With the increasing use of Machine Learning (ML) components in cyber physical systems--so called Learning-Enabled Systems (LESs), there is a recent trend of applying HAZOP-like analysis to LESs. While it shows a great potential to reserve the capability of doing sufficient and systematic safety analysis, there are new technical challenges raised by the novel characteristics of ML that require retrofit of the conventional HAZOP technique. In this regard, we present a new Hierarchical HAZOP-Like method for LESs (HILLS). To deal with the complexity of LESs, HILLS first does "divide and conquer" by stratifying the whole system into three levels, and then proceeds HAZOP on each level to identify (latent-)hazards, causes, security threats and mitigation (with new nodes and guide words). Finally, HILLS attempts at linking and propagating the causal relationship among those identified elements within and across the three levels via both qualitative and quantitative methods. We examine and illustrate the utility of HILLS by a case study on Autonomous Underwater Vehicles, with discussions on assumptions and extensions to real-world applications. HILLS, as a first HAZOP-like attempt on LESs that explicitly considers ML internal behaviours and its interactions with other components, not only uncovers the inherent difficulties of doing safety analysis for LESs, but also demonstrates a good potential to tackle them.

cs.SE↗

arXiv2022

Magnetic Flux Transport Identification of Active Reconnection: MMS Observations in Earth's Magnetosphere

Magnetic reconnection plays an important role in converting energy while modifying field topology. This process takes place in varied plasma environments in which the transport of magnetic flux is intrinsic. Identifying active magnetic reconnection sites in in-situ observations is challenging. A new technique, Magnetic Flux Transport (MFT) analysis, has been developed recently and proven in numerical simulation for identifying active reconnection efficiently and accurately. In this study, we examine the MFT process in 37 previously reported electron diffusion region (EDR)/reconnection-line crossing events at the dayside magnetopause and in the magnetotail and turbulent magnetosheath using Magnetospheric Multiscale measurements. The coexisting inward and outward MFT flows at an X-point provides a signature that magnetic field lines become disconnected and reconnected. The application of MFT analysis to in-situ observations demonstrates that MFT can successfully identify active reconnection sites under complex varied conditions, including asymmetric and turbulent upstream conditions. It also provides a higher rate of identification than plasma outflow jets alone. MFT can be applied to in situ measurements from both single- and multi-spacecraft missions and laboratory experiments.

physics.space-ph↗

arXiv2022

Continual Learning for CTR Prediction: A Hybrid Approach

Click-through rate(CTR) prediction is a core task in cost-per-click(CPC) advertising systems and has been studied extensively by machine learning practitioners. While many existing methods have been successfully deployed in practice, most of them are built upon i.i.d.(independent and identically distributed) assumption, ignoring that the click data used for training and inference is collected through time and is intrinsically non-stationary and drifting. This mismatch will inevitably lead to sub-optimal performance. To address this problem, we formulate CTR prediction as a continual learning task and propose COLF, a hybrid COntinual Learning Framework for CTR prediction, which has a memory-based modular architecture that is designed to adapt, learn and give predictions continuously when faced with non-stationary drifting click data streams. Married with a memory population method that explicitly controls the discrepancy between memory and target data, COLF is able to gain positive knowledge from its historical experience and makes improved CTR predictions. Empirical evaluations on click log collected from a major shopping app in China demonstrate our method's superiority over existing methods. Additionally, we have deployed our method online and observed significant CTR and revenue improvement, which further demonstrates our method's efficacy.

cs.IR↗

arXiv2021

Identification of active magnetic reconnection using magnetic flux transport in plasma turbulence

Magnetic reconnection has been suggested to play an important role in the dynamics and energetics of plasma turbulence by spacecraft observations, simulations and theory over the past two decades, and recently, by magnetosheath observations of MMS. A new method based on magnetic flux transport (MFT) has been developed to identify reconnection activity in turbulent plasmas. This method is applied to a gyrokinetic simulation of two-dimensional (2D) plasma turbulence. Results on the identification of three active reconnection X-points are reported. The first two X-points have developed bi-directional electron outflow jets. Beyond the category of electron-only reconnection, the third X-point does not have bi-directional electron outflow jets because the flow is modified by turbulence. In all cases, this method successfully identifies active reconnection through clear inward and outward flux transport around the X-points. This transport pattern defines reconnection and produces a new quadrupolar structure in the divergence of MFT. This method is expected to be applicable to spacecraft missions such as MMS, Parker Solar Probe, and Solar Orbiter.

physics.plasm-ph↗

arXiv2021

Deep Position-wise Interaction Network for CTR Prediction

Click-through rate (CTR) prediction plays an important role in online advertising and recommender systems. In practice, the training of CTR models depends on click data which is intrinsically biased towards higher positions since higher position has higher CTR by nature. Existing methods such as actual position training with fixed position inference and inverse propensity weighted training with no position inference alleviate the bias problem to some extend. However, the different treatment of position information between training and inference will inevitably lead to inconsistency and sub-optimal online performance. Meanwhile, the basic assumption of these methods, i.e., the click probability is the product of examination probability and relevance probability, is oversimplified and insufficient to model the rich interaction between position and other information. In this paper, we propose a Deep Position-wise Interaction Network (DPIN) to efficiently combine all candidate items and positions for estimating CTR at each position, achieving consistency between offline and online as well as modeling the deep non-linear interaction among position, user, context and item under the limit of serving performance. Following our new treatment to the position bias in CTR prediction, we propose a new evaluation metrics named PAUC (position-wise AUC) that is suitable for measuring the ranking quality at a given position. Through extensive experiments on a real world dataset, we show empirically that our method is both effective and efficient in solving position bias problem. We have also deployed our method in production and observed statistically significant improvement over a highly optimized baseline in a rigorous A/B test.

cs.IR↗

arXiv2020

Universal commensurability augmented Teichmüller space and moduli space

It is known that every finitely unbranched covering $α:\widetilde{S}_{g(α)}\rightarrow S$ of a compact Riemann surface $S$ with genus $g\geq2$ induces an isometric embedding $Γ_α$ from the Teichmüller space $T(S)$ to the Teichüller space $T(\widetilde{S}_{g(α)})$. Actually, it has been showed that the isometric embedding $Γ_α$ can be extended isometrically to the augmented Teichmüller space $\widehat{T}(S)$ of $T(S)$. Using this result, we construct a directed limit $\widehat{T}_{\infty}(S)$ of augmented Teichmüller spaces, where the index runs over all finitely unbranched coverings of $S$. Then, we show that the action of the universal commensurability modular group $Mod_{\infty}(S)$ can extend isometrically on $\widehat{T}_{\infty}(S)$. Furthermore, for any $X_{\infty}\in T_{\infty}(S)$, its orbit of the action of the universal commensurability modular group $Mod_{\infty}(S)$ on the universal commensurability augmented Teichmüller space $\widehat{T}_{\infty}(S)$ is dense. Finally, we also construct a directed limit $\widehat{M}_{\infty}(S)$ of augmented moduli spaces by characteristic towers and show that the subgroup $Caut(π_{1}(S))$ of $Mod_{\infty}(S)$ acts on $\widehat{T}_{\infty}(S)$ to produce $\widehat{M}_{\infty}(S)$ as the quotient.

math.GT↗

arXiv2019

Identifying vital nodes based on reverse greedy method

The identification of vital nodes that maintain the network connectivity is a long-standing challenge in network science. In this paper, we propose a so-called reverse greedy method where the least important nodes are preferentially chosen to make the size of the largest component in the corresponding induced subgraph as small as possible. Accordingly, the nodes being chosen later are more important in maintaining the connectivity. Empirical analyses on ten real networks show that the reverse greedy method performs remarkably better than well-known state-of-the-art methods.

physics.soc-ph↗

arXiv2019

A Deep Spatio-Temporal Fuzzy Neural Network for Passenger Demand Prediction

In spite of its importance, passenger demand prediction is a highly challenging problem, because the demand is simultaneously influenced by the complex interactions among many spatial and temporal factors and other external factors such as weather. To address this problem, we propose a Spatio-TEmporal Fuzzy neural Network (STEF-Net) to accurately predict passenger demands incorporating the complex interactions of all known important factors. We design an end-to-end learning framework with different neural networks modeling different factors. Specifically, we propose to capture spatio-temporal feature interactions via a convolutional long short-term memory network and model external factors via a fuzzy neural network that handles data uncertainty significantly better than deterministic methods. To keep the temporal relations when fusing two networks and emphasize discriminative spatio-temporal feature interactions, we employ a novel feature fusion method with a convolution operation and an attention layer. As far as we know, our work is the first to fuse a deep recurrent neural network and a fuzzy neural network to model complex spatial-temporal feature interactions with additional uncertain input features for predictive learning. Experiments on a large-scale real-world dataset show that our model achieves more than 10% improvement over the state-of-the-art approaches.

cs.LG↗

arXiv2019

Near-optimal capture in the planar circular restricted Pluto-Charon system

In this paper, the near-optimal capture problem is numerically investigated to locate the optimal capture points around Charon using the Pluto-Charon planar circular restricted three-body problem (PCRTBP). Capture orbits are divided into the pre- and post-maneuver portions. In the pre-maneuver portion, the gravitational capture conditions are discussed by backward numerical integration. In the post-maneuver portion, the smaller alignment index (SALI) is applied to numerically investigate the orbital characters of long-term capture orbits around Charon. The initial conditions corresponding to three types of motion: (i) bounded (regular or chaotic), (ii) escaping and (iii) collisional are classified and studied. Combining results of the pre- and post-maneuver portions, the near-optimal capture method is presented to find optimal capture points by overlay figures, and then the corresponding capture orbits are constructed in the PCRTBP. The different results between the Pluto-Charon system and the Earth-Moon system are compared and analyzed. Our results obtained in this paper, including the optimal capture points and the corresponding capture orbits, could be applied in future space mission design.

astro-ph.EP↗

arXiv2018

Multivariate MR Biomarkers Better Predict Cognitive Dysfunction in Mouse Models of Alzheimers Disease

To understand multifactorial conditions such as Alzheimers disease (AD) we need brain signatures that predict the impact of multiple pathologies and their interactions. To help uncover the relationships between brain circuits and cognitive markers we have used mouse models that represent, at least in part, the complex interactions altered in AD. In particular, we aimed to understand the relationship between vulnerable brain circuits and memory deficits measured in the Morris water maze, and we tested several predictive modeling approaches. We used in vivo manganese enhanced MRI voxel based analyses to reveal regional differences in volume (morphometry), signal intensity (activity), and magnetic susceptibility (iron deposition, demyelination). These regions included the hippocampus, olfactory areas, entorhinal cortex and cerebellum. The image based properties of these regions were used to predict spatial memory. We next used eigenanatomy, which reduces dimensionality to produce sets of regions that explain the variance in the data. For each imaging marker, eigenanatomy revealed networks underpinning a range of cognitive functions including memory, motor function, and associative learning. Finally, the integration of multivariate markers in a supervised sparse canonical correlation approach outperformed single predictor models and had significant correlates to spatial memory. Among a priori selected regions, the fornix also provided good predictors, raising the possibility of investigating how disease propagation within brain networks leads to cognitive deterioration. Our results support that modeling approaches integrating multivariate imaging markers provide sensitive predictors of AD-like behaviors. Such strategies for mapping brain circuits responsible for behaviors may help in the future predict disease progression, or response to interventions.

q-bio.QM↗

arXiv2017

Consensus in Self-similar Hierarchical Graphs and Sierpiński Graphs: Convergence Speed, Delay Robustness, and Coherence

The hierarchical graphs and Sierpiński graphs are constructed iteratively, which have the same number of vertices and edges at any iteration, but exhibit quite different structural properties: the hierarchical graphs are non-fractal and small-world, while the Sierpiński graphs are fractal and "large-world". Both graphs have found broad applications. In this paper, we study consensus problems in hierarchical graphs and Sierpiński graphs, focusing on three important quantities of consensus problems, that is, convergence speed, delay robustness, and coherence for first-order (and second-order) dynamics, which are, respectively, determined by algebraic connectivity, maximum eigenvalue, and sum of reciprocal (and square of reciprocal) of each nonzero eigenvalue of Laplacian matrix. For both graphs, based on the explicit recursive relation of eigenvalues at two successive iterations, we evaluate the second smallest eigenvalue, as well as the largest eigenvalue, and obtain the closed-form solutions to the sum of reciprocals (and square of reciprocals) of all nonzero eigenvalues. We also compare our obtained results for consensus problems on both graphs and show that they differ in all quantities concerned, which is due to the marked difference of their topological structures.

eess.SY↗

arXiv2017

Extended corona product as an exactly tractable model for weighted heterogeneous networks

Various graph products and operations have been widely used to construct complex networks with common properties of real-life systems. However, current works mainly focus on designing models of binary networks, in spite of the fact that many real networks can be better mimicked by heterogeneous weighted networks. In this paper, we develop a corona product of two weighted graphs, based on which and an observed updating mechanism of edge weight in real networks, we propose a minimal generative model for inhomogeneous weighted networks. We derive analytically relevant properties of the weighted network model, including strength, weight and degree distributions, clustering coefficient, degree correlations and diameter. These properties are in good agreement with those observed in diverse real-world weighted networks. We then determine all the eigenvalues and their corresponding multiplicities of the transition probability matrix for random walks on the weighted networks. Finally, we apply the obtained spectra to derive explicit expressions for mean hitting time of random walks and weighted counting of spanning trees on the weighted networks. Our model is an exactly solvable one, allowing to analytically treat its structural and dynamical properties, which is thus a good test-bed and an ideal substrate network for studying different dynamical processes, in order to explore the impacts of heterogeneous weight distribution on these processes.

cs.SI↗

arXiv2010

Explicit determination of mean first-passage time for random walks on deterministic uniform recursive trees

The determination of mean first-passage time (MFPT) for random walks in networks is a theoretical challenge, and is a topic of considerable recent interest within the physics community. In this paper, according to the known connections between MFPT, effective resistance, and the eigenvalues of graph Laplacian, we first study analytically the MFPT between all node pairs of a class of growing treelike networks, which we term deterministic uniform recursive trees (DURTs), since one of its particular cases is a deterministic version of the famous uniform recursive tree. The interesting quantity is determined exactly through the recursive relation of the Laplacian spectra obtained from the special construction of DURTs. The analytical result shows that the MFPT between all couples of nodes in DURTs varies as $N \ln N$ for large networks with node number $N$. Second, we study trapping on a particular network of DURTs, focusing on a special case with the immobile trap positioned at a node having largest degree. We determine exactly the average trapping time (ATT) that is defined as the average of FPT from all nodes to the trap. In contrast to the scaling of the MFPT, the leading behavior of ATT is a linear function of $N$. Interestingly, we show that the behavior for ATT of the trapping problem is related to the trapping location, which is in comparison with the phenomenon of trapping on fractal T-graph although both networks exhibit treestructure. Finally, we believe that the methods could open the way to exactly calculate the MFPT and ATT in a wide range of deterministic media.

cond-mat.stat-mech↗

arXiv2009

Recursive solutions for Laplacian spectra and eigenvectors of a class of growing treelike networks

The complete knowledge of Laplacian eigenvalues and eigenvectors of complex networks plays an outstanding role in understanding various dynamical processes running on them; however, determining analytically Laplacian eigenvalues and eigenvectors is a theoretical challenge. In this paper, we study the Laplacian spectra and their corresponding eigenvectors of a class of deterministically growing treelike networks. The two interesting quantities are determined through the recurrence relations derived from the structure of the networks. Beginning from the rigorous relations one can obtain the complete eigenvalues and eigenvectors for the networks of arbitrary size. The analytical method opens the way to analytically compute the eigenvalues and eigenvectors of some other deterministic networks, making it possible to accurately calculate their spectral characteristics.

cond-mat.stat-mech↗

arXiv2009

Structural and spectral properties of a family of deterministic recursive trees: Rigorous solutions

As one of the most significant models, the uniform recursive tree (URT) has found many applications in a variety of fields. In this paper, we study rigorously the structural features and spectral properties of the adjacency matrix for a family of deterministic uniform recursive trees (DURTs) that are deterministic versions of URT. Firstly, from the perspective of complex networks, we investigate analytically the main structural characteristics of DURTs, and obtain the accurate solutions for these properties, which include degree distribution, average path length, distribution of node betweenness, and degree correlations. Then we determine the complete eigenvalues and their corresponding eigenvectors of the adjacency matrix for DURTs. Our research may shed light in better understanding of the features for URT. Also, the analytical methods used here is capable of extending to many other deterministic networks, making the precise computation of their properties (especially the full spectrum characteristics) possible.

cond-mat.stat-mech↗

arXiv2009

Random walks on the Apollonian network with a single trap

Explicit determination of the mean first-passage time (MFPT) for trapping problem on complex media is a theoretical challenge. In this paper, we study random walks on the Apollonian network with a trap fixed at a given hub node (i.e. node with the highest degree), which are simultaneously scale-free and small-world. We obtain the precise analytic expression for the MFPT that is confirmed by direct numerical calculations. In the large system size limit, the MFPT approximately grows as a power-law function of the number of nodes, with the exponent much less than 1, which is significantly different from the scaling for some regular networks or fractals, such as regular lattices, Sierpinski fractals, T-graph, and complete graphs. The Apollonian network is the most efficient configuration for transport by diffusion among all previously studied structure.

cond-mat.stat-mech↗

arXiv2009

Exact solution for mean first-passage time on a pseudofractal scale-free web

The explicit determinations of the mean first-passage time (MFPT) for trapping problem are limited to some simple structure, e.g., regular lattices and regular geometrical fractals, and determining MFPT for random walks on other media, especially complex real networks, is a theoretical challenge. In this paper, we investigate a simple random walk on the the pseudofractal scale-free web (PSFW) with a perfect trap located at a node with the highest degree, which simultaneously exhibits the remarkable scale-free and small-world properties observed in real networks. We obtain the exact solution for the MFPT that is calculated through the recurrence relations derived from the structure of PSFW. The rigorous solution exhibits that the MFPT approximately increases as a power-law function of the number of nodes, with the exponent less than 1. We confirm the closed-form solution by direct numerical calculations. We show that the structure of PSFW can improve the efficiency of transport by diffusion, compared with some other structure, such as regular lattices, Sierpinski fractals, and T-graph. The analytical method can be applied to other deterministic networks, making the accurate computation of MFPT possible.

cond-mat.stat-mech↗

arXiv2008

Topologies and Laplacian spectra of a deterministic uniform recursive tree

The uniform recursive tree (URT) is one of the most important models and has been successfully applied to many fields. Here we study exactly the topological characteristics and spectral properties of the Laplacian matrix of a deterministic uniform recursive tree, which is a deterministic version of URT. Firstly, from the perspective of complex networks, we determine the main structural characteristics of the deterministic tree. The obtained vigorous results show that the network has an exponential degree distribution, small average path length, power-law distribution of node betweenness, and positive degree-degree correlations. Then we determine the complete Laplacian spectra (eigenvalues) and their corresponding eigenvectors of the considered graph. Interestingly, all the Laplacian eigenvalues are distinct.

cond-mat.stat-mech↗