arXiv subjects

Shenghua Liu

Publications and source records attributed to Shenghua Liu.

54 records · Page 3Linked to original sources

arXiv2023

Complex strengthening mechanisms in nanocrystalline Ni-Mo alloys revealed by a machine-learning interatomic potential

A nanocrystalline metal's strength increases significantly as its grain size decreases, a phenomenon known as the Hall-Petch relation. Such relation, however, breaks down when the grains become too small. Experimental studies have circumvented this problem in a set of Ni-Mo alloys by stabilizing the grain boundaries (GB). Here, using atomistic simulations with a machine learning-based interatomic potential (ML-IAP), we demonstrate that the inverse Hall-Petch relation can be correctly reproduced due to a change in the dominant deformation mechanism as the grain becomes small in the Ni-Mo polycrystals. It is found that the atomic von Mises strain can be significantly reduced by either solute doping and/or annealing for small-grain-size polycrystals, leading to the increased strength of the polycrystals. On the other hand, for large-grain-size polycrystals, annealing weakens the material due to the large atomic movements in GB regions. Over a broad range of grain size, the superposition of the solute and annealing effects on polycrystals enhances the strength of those with small grain size more than those with large ones, giving rise to the continuous strengthening at extremely small grain sizes. Overall, this study not only demonstrates the reliability of the ML-IAP, but also provides a comprehensive atomistic view of complex strengthening mechanisms in nanocrystals, opening up a new avenue to tailor their mechanical properties.

cond-mat.mtrl-sci↗

arXiv2022

PRISM: A Non-Equilibrium, Multiphase Interstellar Medium Model for Radiation Hydrodynamics Simulations of Galaxies

We introduce the PRISM interstellar medium (ISM) model for thermochemistry and its implementation in the RAMSES-RTZ code. The model includes a non-equilibrium primordial, metal, and molecular chemistry network for 115 species coupled to on-the-fly multifrequency radiation transport. PRISM accurately accounts for the dominant ISM cooling and heating processes in the low-density regime (i.e. $ρ<10^5\ {\rm cm^{-3}}$), including photoheating, photoelectric heating, H$_2$ heating/cooling, cosmic-ray heating, H/He cooling, metal-line cooling, CO cooling, and dust cooling (recombination and gas-grain collisions). We validate the model by comparing 1D equilibrium simulations across six dex in metallicity to existing 1D ISM models in the literature. We apply PRISM to high-resolution (4.5 pc) isolated dwarf galaxy simulations that include state-of-the-art models for star formation and stellar feedback to take an inventory of which cooling and heating processes dominate each different gas phase of a galaxy and to understand the importance of non-equilibrium effects. We show that most of the ISM gas is either close to thermal equilibrium or exhibits a slight cooling instability, while from a chemical perspective, the non-equilibrium electron fraction is often more than three times higher or lower than the equilibrium value, which impacts cooling, heating, and observable emission lines. Electron enhancements are attributed to recombination lags while deficits are shown to be due to rapid cosmic-ray heating. The PRISM model and its coupling to RAMSES-RTZ is applicable to a wide variety of astrophysical scenarios, from cosmological simulations to isolated giant molecular clouds, and is particularly useful for understanding how changes to ISM physics impact observable quantities such as metallic emission lines.

astro-ph.GA↗

arXiv2022

Learning node embeddings via summary graphs: a brief theoretical analysis

Graph representation learning plays an important role in many graph mining applications, but learning embeddings of large-scale graphs remains a problem. Recent works try to improve scalability via graph summarization -- i.e., they learn embeddings on a smaller summary graph, and then restore the node embeddings of the original graph. However, all existing works depend on heuristic designs and lack theoretical analysis. Different from existing works, we contribute an in-depth theoretical analysis of three specific embedding learning methods based on introduced kernel matrix, and reveal that learning embeddings via graph summarization is actually learning embeddings on a approximate graph constructed by the configuration model. We also give analysis about approximation error. To the best of our knowledge, this is the first work to give theoretical analysis of this approach. Furthermore, our analysis framework gives interpretation of some existing methods and provides great insights for future work on this problem.

cs.LG↗

arXiv2022

Multi-scale Anomaly Detection for Big Time Series of Industrial Sensors

Given a multivariate big time series, can we detect anomalies as soon as they occur? Many existing works detect anomalies by learning how much a time series deviates away from what it should be in the reconstruction framework. However, most models have to cut the big time series into small pieces empirically since optimization algorithms cannot afford such a long series. The question is raised: do such cuts pollute the inherent semantic segments, like incorrect punctuation in sentences? Therefore, we propose a reconstruction-based anomaly detection method, MissGAN, iteratively learning to decode and encode naturally smooth time series in coarse segments, and finding out a finer segment from low-dimensional representations based on HMM. As a result, learning from multi-scale segments, MissGAN can reconstruct a meaningful and robust time series, with the help of adversarial regularization and extra conditional states. MissGAN does not need labels or only needs labels of normal instances, making it widely applicable. Experiments on industrial datasets of real water network sensors show our MissGAN outperforms the baselines with scalability. Besides, we use a case study on the CMU Motion dataset to demonstrate that our model can well distinguish unexpected gestures from a given conditional motion.

cs.LG↗

arXiv2022

MonLAD: Money Laundering Agents Detection in Transaction Streams

Given a stream of money transactions between accounts in a bank, how can we accurately detect money laundering agent accounts and suspected behaviors in real-time? Money laundering agents try to hide the origin of illegally obtained money by dispersive multiple small transactions and evade detection by smart strategies. Therefore, it is challenging to accurately catch such fraudsters in an unsupervised manner. Existing approaches do not consider the characteristics of those agent accounts and are not suitable to the streaming settings. Therefore, we propose MonLAD and MonLAD-W to detect money laundering agent accounts in a transaction stream by keeping track of their residuals and other features; we devise AnoScore algorithm to find anomalies based on the robust measure of statistical deviation. Experimental results show that MonLAD outperforms the state-of-the-art baselines on real-world data and finds various suspicious behavior patterns of money laundering. Additionally, several detected suspected accounts have been manually-verified as agents in real money laundering scenario.

cs.SI↗

arXiv2021

AugSplicing: Synchronized Behavior Detection in Streaming Tensors

How can we track synchronized behavior in a stream of time-stamped tuples, such as mobile devices installing and uninstalling applications in the lockstep, to boost their ranks in the app store? We model such tuples as entries in a streaming tensor, which augments attribute sizes in its modes over time. Synchronized behavior tends to form dense blocks (i.e. subtensors) in such a tensor, signaling anomalous behavior, or interesting communities. However, existing dense block detection methods are either based on a static tensor, or lack an efficient algorithm in a streaming setting. Therefore, we propose a fast streaming algorithm, AugSplicing, which can detect the top dense blocks by incrementally splicing the previous detection with the incoming ones in new tuples, avoiding re-runs over all the history data at every tracking time step. AugSplicing is based on a splicing condition that guides the algorithm (Section 4). Compared to the state-of-the-art methods, our method is (1) effective to detect fraudulent behavior in installing data of real-world apps and find a synchronized group of students with interesting features in campus Wi-Fi data; (2) robust with splicing theory for dense block detection; (3) streaming and faster than the existing streaming algorithm, with closely comparable accuracy.

cs.DB↗

arXiv2021

CubeFlow: Money Laundering Detection with Coupled Tensors

Money laundering (ML) is the behavior to conceal the source of money achieved by illegitimate activities, and always be a fast process involving frequent and chained transactions. How can we detect ML and fraudulent activity in large scale attributed transaction data (i.e.~tensors)? Most existing methods detect dense blocks in a graph or a tensor, which do not consider the fact that money are frequently transferred through middle accounts. CubeFlow proposed in this paper is a scalable, flow-based approach to spot fraud from a mass of transactions by modeling them as two coupled tensors and applying a novel multi-attribute metric which can reveal the transfer chains accurately. Extensive experiments show CubeFlow outperforms state-of-the-art baselines in ML behavior detection in both synthetic and real data.

cs.CY↗

arXiv2020

Summarizing graphs using the configuration model

Given a large graph, how can we summarize it with fewer nodes and edges while maintaining its key properties, such as spectral property? Although graphs play more and more important roles in many real-world applications, the growth of their size presents great challenges to graph analysis. As a solution, graph summarization, which aims to find a compact representation that preserves the important properties of a given graph, has received much attention, and numerous algorithms have been developed for it. However, most of the algorithms adopt the uniform reconstruction scheme, which is based on an unrealistic assumption that edges are uniformly distributed. In this work, we propose a novel and realistic reconstruction scheme, which preserves the degree of nodes, and we develop an efficient graph summarization algorithm called DPGS based on the Minimum Description Length principle. We theoretically analyze the difference between the original and summary graphs from a spectral perspective, and we perform extensive experiments on multiple real-world datasets. The results show that DPGS yields compact representation that preserves the essential properties of the original graph.

cs.SI↗

arXiv2020

Efficient Model-Based Collaborative Filtering with Fast Adaptive PCA

A model-based collaborative filtering (CF) approach utilizing fast adaptive randomized singular value decomposition (SVD) is proposed for the matrix completion problem in recommender system. Firstly, a fast adaptive PCA frameworkis presented which combines the fixed-precision randomized matrix factorization algorithm [1] and accelerating skills for handling large sparse data. Then, a novel termination mechanism for the adaptive PCA is proposed to automatically determine a number of latent factors for achieving the near optimal prediction accuracy during the subsequent model-based CF. The resulted CF approach has good accuracy while inheriting high runtime efficiency. Experiments on real data show that, the proposed adaptive PCA is up to 2.7X and 6.7X faster than the original fixed-precision SVD approach [1] and svds in Matlab repsectively, while preserving accuracy. The proposed model-based CF approach is able to efficiently process the MovieLens data with 20M ratings and exhibits more than 10X speedup over the regularized matrix factorization based approach [2] and the fast singular value thresholding approach [3] with comparable or better accuracy. It also owns the advantage of parameter free. Compared with the deep-learning-based CF approach, the proposed approach is much more computationally efficient, with just marginal performance loss.

cs.LG↗

arXiv2018

Optimal Algorithm for Profiling Dynamic Arrays with Finite Values

How can one quickly answer the most and top popular objects at any time, given a large log stream in a system of billions of users? It is equivalent to find the mode and top-frequent elements in a dynamic array corresponding to the log stream. However, most existing work either restrain the dynamic array within a sliding window, or do not take advantages of only one element can be added or removed in a log stream. Therefore, we propose a profiling algorithm, named S-Profile, which is of $O(1)$ time complexity for every updating of the dynamic array, and optimal in terms of computational complexity. With the profiling results, answering the queries on the statistics of dynamic array becomes trivial and fast. With the experiments of various settings of dynamic arrays, our accurate S-Profile algorithm outperforms the well-known methods, showing at least 2X speedup to the heap based approach and 13X or larger speedup to the balanced tree based approach.

cs.DS↗

arXiv2017

EagleMine: Vision-Guided Mining in Large Graphs

Given a graph with millions of nodes, what patterns exist in the distributions of node characteristics, and how can we detect them and separate anomalous nodes in a way similar to human vision? In this paper, we propose a vision-guided algorithm, EagleMine, to summarize micro-cluster patterns in two-dimensional histogram plots constructed from node features in a large graph. EagleMine utilizes a water-level tree to capture cluster structures according to vision-based intuition at multi-resolutions. EagleMine traverses the water-level tree from the root and adopts statistical hypothesis tests to determine the optimal clusters that should be fitted along the path, and summarizes each cluster with a truncated Gaussian distribution. Experiments on real data show that our method can find truncated and overlapped elliptical clusters, even when some baseline methods split one visual cluster into pieces with Gaussian spheres. To identify potentially anomalous microclusters, EagleMine also a designates score to measure the suspiciousness of outlier groups (i.e. node clusters) and outlier nodes, detecting bots and anomalous users with high accuracy in the real Microblog data.

cs.SI↗

arXiv2017

HoloScope: Topology-and-Spike Aware Fraud Detection

As online fraudsters invest more resources, including purchasing large pools of fake user accounts and dedicated IPs, fraudulent attacks become less obvious and their detection becomes increasingly challenging. Existing approaches such as average degree maximization suffer from the bias of including more nodes than necessary, resulting in lower accuracy and increased need for manual verification. Hence, we propose HoloScope, which uses information from graph topology and temporal spikes to more accurately detect groups of fraudulent users. In terms of graph topology, we introduce "contrast suspiciousness," a dynamic weighting approach, which allows us to more accurately detect fraudulent blocks, particularly low-density blocks. In terms of temporal spikes, HoloScope takes into account the sudden bursts and drops of fraudsters' attacking patterns. In addition, we provide theoretical bounds for how much this increases the time cost needed for fraudsters to conduct adversarial attacks. Additionally, from the perspective of ratings, HoloScope incorporates the deviation of rating scores in order to catch fraudsters more accurately. Moreover, HoloScope has a concise framework and sub-quadratic time complexity, making the algorithm reproducible and scalable. Extensive experiments showed that HoloScope achieved significant accuracy improvements on synthetic and real data, compared with state-of-the-art fraud detection methods.

cs.SI↗

arXiv2017

Learning Sentimental Influences from Users' Behaviors

Modeling interpersonal influence on different sentimental polarities is a fundamental problem in opinion formation and viral marketing. There has not been seen an effective solution for learning sentimental influences from users' behaviors yet. Previous related works on information propagation directly define interpersonal influence between each pair of users as a parameter, which is independent from each others, even if the influences come from or affect the same user. And influences are learned from user's propagation behaviors, namely temporal cascades, while sentiments are not associated with them. Thus we propose to model the interpersonal influence by latent influence and susceptibility matrices defined on individual users and sentiment polarities. Such low-dimensional and distributed representations naturally make the interpersonal influences related to the same user coupled with each other, and in turn, reduce the model complexity. Sentiments act on different rows of parameter matrices, depicting their effects in modeling cascades. With the iterative optimization algorithm of projected stochastic gradient descent over shuffled mini-batches and Adadelta update rule, negative cases are repeatedly sampled with the distribution of infection frequencies users, for reducing computation cost and optimization imbalance. Experiments are conducted on Microblog dataset. The results show that our model achieves better performance than the state-of-the-art and pair-wise models. Besides, analyzing the distribution of learned users' sentimental influences and susceptibilities results some interesting discoveries.

cs.SI↗

arXiv2017

Single-Pass PCA of Large High-Dimensional Data

Principal component analysis (PCA) is a fundamental dimension reduction tool in statistics and machine learning. For large and high-dimensional data, computing the PCA (i.e., the singular vectors corresponding to a number of dominant singular values of the data matrix) becomes a challenging task. In this work, a single-pass randomized algorithm is proposed to compute PCA with only one pass over the data. It is suitable for processing extremely large and high-dimensional data stored in slow memory (hard disk) or the data generated in a streaming fashion. Experiments with synthetic and real data validate the algorithm's accuracy, which has orders of magnitude smaller error than an existing single-pass algorithm. For a set of high-dimensional data stored as a 150 GB file, the proposed algorithm is able to compute the first 50 principal components in just 24 minutes on a typical 24-core computer, with less than 1 GB memory cost.

cs.DS↗

arXiv2017

Marked Temporal Dynamics Modeling based on Recurrent Neural Network

We are now witnessing the increasing availability of event stream data, i.e., a sequence of events with each event typically being denoted by the time it occurs and its mark information (e.g., event type). A fundamental problem is to model and predict such kind of marked temporal dynamics, i.e., when the next event will take place and what its mark will be. Existing methods either predict only the mark or the time of the next event, or predict both of them, yet separately. Indeed, in marked temporal dynamics, the time and the mark of the next event are highly dependent on each other, requiring a method that could simultaneously predict both of them. To tackle this problem, in this paper, we propose to model marked temporal dynamics by using a mark-specific intensity function to explicitly capture the dependency between the mark and the time of the next event. Extensive experiments on two datasets demonstrate that the proposed method outperforms state-of-the-art methods at predicting marked temporal dynamics.

cs.LG↗

arXiv2016

Efficient Thermal Conductance in Organometallic Perovskite CH3NH3PbI3 Films

Perovskite-based optoelectronic devices have shown great promise for solar conversion and other optoelectronic applications, but their long-term performance instability is regarded as a major obstacle to their widespread deployment. Previous works have shown that the ultralow thermal conductivity and inefficient heat spreading might put an intrinsic limit on the lifetime of perovskite devices. Here, we report the observation of a remarkably efficient thermal conductance, with conductivity of 11.2 +/- 0.8 W m^-1 K^-1 at room temperature, in densely-packed perovskite CH3NH3PbI3 films, via noncontact time-domain thermal reflectance measurements. The temperature-dependent experiments suggest the important roles of organic cations and structural phase transitions, which are further confirmed by temperature-dependent Raman spectra. The thermal conductivity at room temperature observed here is over one order of magnitude larger than that in the early report, suggesting that perovskite device performance will not be limited by thermal stability.

cond-mat.mtrl-sci↗

arXiv2015

Learning user-specific latent influence and susceptibility from information cascades

Predicting cascade dynamics has important implications for understanding information propagation and launching viral marketing. Previous works mainly adopt a pair-wise manner, modeling the propagation probability between pairs of users using n^2 independent parameters for n users. Consequently, these models suffer from severe overfitting problem, specially for pairs of users without direct interactions, limiting their prediction accuracy. Here we propose to model the cascade dynamics by learning two low-dimensional user-specific vectors from observed cascades, capturing their influence and susceptibility respectively. This model requires much less parameters and thus could combat overfitting problem. Moreover, this model could naturally model context-dependent factors like cumulative effect in information propagation. Extensive experiments on synthetic dataset and a large-scale microblogging dataset demonstrate that this model outperforms the existing pair-wise models at predicting cascade dynamics, cascade size, and "who will be retweeted".

cs.SI↗

arXiv2013

Orbital-Selective Mottness in KxFe2-ySe2 Superconductors Revealed by Pump-Probe Spectroscopy

We report transient optical signatures of the orbital-selective Mottness in superconducting KxFe2-ySe2 crystals by using dual-color pump-probe spectroscopy. Besides multi-exponential decay recovery dynamics of photo-induced quasiparticles, a damped oscillatory component due to coherent acoustic phonons emerges when the superconducting phase is suppressed by increasing the temperature or excitation power. The oscillatory component diminishes with significant enhancement of a slow decay component upon raising temperature to 150-160 K. These results are in consistence with the picture of orbital-selective Mott phase transition, indicating a vital role played by electron correlation in the iron-based superconductors.

cond-mat.supr-con↗