arXiv subjects

Xuan Zhang

Publications and source records attributed to Xuan Zhang.

At least 181 records · Page 10Linked to original sources

arXiv2022

A Fair and Efficient Hybrid Federated Learning Framework based on XGBoost for Distributed Power Prediction

In a modern power system, real-time data on power generation/consumption and its relevant features are stored in various distributed parties, including household meters, transformer stations and external organizations. To fully exploit the underlying patterns of these distributed data for accurate power prediction, federated learning is needed as a collaborative but privacy-preserving training scheme. However, current federated learning frameworks are polarized towards addressing either the horizontal or vertical separation of data, and tend to overlook the case where both are present. Furthermore, in mainstream horizontal federated learning frameworks, only artificial neural networks are employed to learn the data patterns, which are considered less accurate and interpretable compared to tree-based models on tabular datasets. To this end, we propose a hybrid federated learning framework based on XGBoost, for distributed power prediction from real-time external features. In addition to introducing boosted trees to improve accuracy and interpretability, we combine horizontal and vertical federated learning, to address the scenario where features are scattered in local heterogeneous parties and samples are scattered in various local districts. Moreover, we design a dynamic task allocation scheme such that each party gets a fair share of information, and the computing power of each party can be fully leveraged to boost training efficiency. A follow-up case study is presented to justify the necessity of adopting the proposed framework. The advantages of the proposed framework in fairness, efficiency and accuracy performance are also confirmed.

cs.LG↗

arXiv2021

Data-Driven Controllability Analysis and Stabilization for Linear Descriptor Systems

For a parameter-unknown linear descriptor system, this paper proposes data-driven methods to testify the system's type and controllability and then to stabilize it. First, a data-based condition is developed to identify whether this unknown system is a descriptor system or is equivalent to a normal system. Furthermore, various controllability concepts are testified by replacing the descriptor system's matrices with data. Finally, a data-based decomposing method is proposed to transfer the nominal system into its slow-fast subsystems' form, so that a state feedback controller for the slow subsystem can be obtained from persistently exciting input and state sequences. Meanwhile, due to the equivalent stabilizability between the nominal system and its slow subsystem, a state feedback controller which stabilizes the nominal system is also obtained. A simulation example is provided to illustrate the effectiveness of those methods.

eess.SY↗

arXiv2021

A Federated Learning Framework for Smart Grids: Securing Power Traces in Collaborative Learning

With the deployment of smart sensors and advancements in communication technologies, big data analytics have become vastly popular in the smart grid domain, informing stakeholders of the best power utilization strategy. However, these power-related data are stored and owned by different parties. For example, power consumption data are stored in numerous transformer stations across cities; mobility data of the population, which are important indicators of power consumption, are held by mobile companies. Direct data sharing might compromise party benefits, individual privacy and even national security. Inspired by the federated learning scheme from Google AI, we propose a federated learning framework for smart grids, which enables collaborative learning of power consumption patterns without leaking individual power traces. Horizontal federated learning is employed when data are scattered in the sample space; vertical federated learning, on the other hand, is designed for the case with data scattered in the feature space. Case studies show that, with proper encryption schemes such as Paillier encryption, the machine learning models constructed from the proposed framework are lossless, privacy-preserving and effective. Finally, the promising future of federated learning in other facets of the smart grid is discussed, including electric vehicles, distributed generation/consumption and integrated energy systems.

cs.LG↗

arXiv2021

Flow states and heat transport in Rayleigh--Bénard convection with different sidewall boundary conditions

This work addresses the effects of different thermal sidewall boundary conditions on the formation of flow states and heat transport in two- and three-dimensional Rayleigh--Bénard convection (RBC) by means of direct numerical simulations and steady-state analysis for Rayleigh numbers $Ra$ up to $4\times10^{10}$ and Prandtl numbers $Pr=0.1,1$ and $10$. We show that a linear temperature profile imposed at the conductive sidewall leads to a premature collapse of the single-roll state, whereas a sidewall maintained at a constant temperature enhances its stability. The collapse is caused by accelerated growth of the corner rolls with two distinct growth rate regimes determined by diffusion or convection for small or large $Ra$, respectively. Above the collapse of the single-roll state, we find the emergence of a double-roll state in two-dimensional RBC and a double-toroidal state in three-dimensional cylindrical RBC. These states are most prominent in RBC with conductive sidewalls. The different states are reflected in the global heat transport, so that the different thermal conditions at the sidewall lead to significant differences in the Nusselt number for small to moderate $Ra$. However, for larger $Ra$, heat transport and flow dynamics become increasingly alike for different sidewalls and are almost indistinguishable for $Ra>10^9$. This suggests that the influence of imperfectly insulated sidewalls in RBC experiments is insignificant at very high $Ra$ - provided that the mean sidewall temperature is controlled.

physics.flu-dyn↗

arXiv2021

A Sequence to Sequence Model for Extracting Multiple Product Name Entities from Dialog

E-commerce voice ordering systems need to recognize multiple product name entities from ordering utterances. Existing voice ordering systems such as Amazon Alexa can capture only a single product name entity. This restrains users from ordering multiple items with one utterance. In recent years, pre-trained language models, e.g., BERT and GPT-2, have shown promising results on NLP benchmarks like Super-GLUE. However, they can't perfectly generalize to this Multiple Product Name Entity Recognition (MPNER) task due to the ambiguity in voice ordering utterances. To fill this research gap, we propose Entity Transformer (ET) neural network architectures which recognize up to 10 items in an utterance. In our evaluation, the best ET model (conveRT + ngram + ET) has a performance improvement of 12% on our test set compared to the non-neural model, and outperforms BERT with ET as well. This helps customers finalize their shopping cart via voice dialog, which improves shopping efficiency and experience.

cs.CL↗

arXiv2021

Millimeter-sized Dust Grains Appear Surviving the Water-sublimating Temperature in the Inner 10 au of the FU Ori Disk

Previous observations have shown that the $\lesssim$10 au, $\gtrsim$400 K hot inner disk of the archetypal accretion outburst young stellar object, FU Ori, is dominated by viscous heating. To constrain dust properties in this region, we have performed radio observations toward this disk using the Karl G. Jansky Very Large Array (JVLA) in 2020 June-July, September, and November. We also performed complementary optical photometric monitoring observations. We found that the dust thermal emission from the hot inner disk mid-plane of FU Ori has been approximately stationary and the maximum dust grain size is $\gtrsim$1.6 mm in this region. If the hot inner disk of FU Ori which is inward of the 150-170 K water snowline is turbulent (e.g., corresponding to a Sunyaev & Shakura viscous $α_{t}\gtrsim$0.1), or if the actual maximum grain size is still larger than the lower limit we presently constrain, then as suggested by the recent analytical calculations and the laboratory measurements, water-ice free dust grains may be stickier than water-ice coated dust grains in protoplanetary disks. Additionally, we find that the free-free emission and the Johnson B and V bands magnitudes of these binary stars are brightening in 2016-2020. The optical and radio variability might be related to the dynamically evolving protostellar or disk accretion activities. Our results highlight that hot inner disks of outbursting objects are important laboratories for testing models of dust grain growth. Given the active nature of such systems, to robustly diagnose the maximum dust grain sizes, it is important to carry out coordinated multi-wavelength radio observations.

astro-ph.EP↗

arXiv2021

On the Convergence of Tsetlin Machines for the IDENTITY- and NOT Operators

The Tsetlin Machine (TM) is a recent machine learning algorithm with several distinct properties, such as interpretability, simplicity, and hardware-friendliness. Although numerous empirical evaluations report on its performance, the mathematical analysis of its convergence is still open. In this article, we analyze the convergence of the TM with only one clause involved for classification. More specifically, we examine two basic logical operators, namely, the "IDENTITY"- and "NOT" operators. Our analysis reveals that the TM, with just one clause, can converge correctly to the intended logical operator, learning from training data over an infinite time horizon. Besides, it can capture arbitrarily rare patterns and select the most accurate one when two candidate patterns are incompatible, by configuring a granularity parameter. The analysis of the convergence of the two basic operators lays the foundation for analyzing other logical operators. These analyses altogether, from a mathematical perspective, provide new insights on why TMs have obtained state-of-the-art performance on several pattern recognition problems.

cs.AI↗

arXiv2021

DIG: A Turnkey Library for Diving into Graph Deep Learning Research

Although there exist several libraries for deep learning on graphs, they are aiming at implementing basic operations for graph deep learning. In the research community, implementing and benchmarking various advanced tasks are still painful and time-consuming with existing libraries. To facilitate graph deep learning research, we introduce DIG: Dive into Graphs, a turnkey library that provides a unified testbed for higher level, research-oriented graph deep learning tasks. Currently, we consider graph generation, self-supervised learning on graphs, explainability of graph neural networks, and deep learning on 3D graphs. For each direction, we provide unified implementations of data interfaces, common algorithms, and evaluation metrics. Altogether, DIG is an extensible, open-source, and turnkey library for researchers to develop new methods and effortlessly compare with common baselines using widely used datasets and evaluation metrics. Source code is available at https://github.com/divelab/DIG.

cs.LG↗

arXiv2021

Molecule3D: A Benchmark for Predicting 3D Geometries from Molecular Graphs

Graph neural networks are emerging as promising methods for modeling molecular graphs, in which nodes and edges correspond to atoms and chemical bonds, respectively. Recent studies show that when 3D molecular geometries, such as bond lengths and angles, are available, molecular property prediction tasks can be made more accurate. However, computing of 3D molecular geometries requires quantum calculations that are computationally prohibitive. For example, accurate calculation of 3D geometries of a small molecule requires hours of computing time using density functional theory (DFT). Here, we propose to predict the ground-state 3D geometries from molecular graphs using machine learning methods. To make this feasible, we develop a benchmark, known as Molecule3D, that includes a dataset with precise ground-state geometries of approximately 4 million molecules derived from DFT. We also provide a set of software tools for data processing, splitting, training, and evaluation, etc. Specifically, we propose to assess the error and validity of predicted geometries using four metrics. We implement two baseline methods that either predict the pairwise distance between atoms or atom coordinates in 3D space. Experimental results show that, compared with generating 3D geometries with RDKit, our method can achieve comparable prediction accuracy but with much smaller computational costs. Our Molecule3D is available as a module of the MoleculeX software library (https://github.com/divelab/MoleculeX).

cs.LG↗

arXiv2021

Connecting wall modes and boundary zonal flows in rotating Rayleigh--Bénard convection

Using direct numerical simulations, we study rotating Rayleigh-Bénard convection in a cylindrical cell for a broad range of Rayleigh, Ekman, and Prandtl numbers from the onset of wall modes to the geostrophic regime, an extremely important one in geophysical and astrophysical contexts. We connect linear wall-mode states that occur prior to the onset of bulk convection with the boundary zonal flow that coexists with turbulent bulk convection in the geostrophic regime through the continuity of length and time scales and of convective heat transport. We quantitatively collapse drift frequency, boundary length, and heat transport data from numerous sources over many orders of magnitude in Rayleigh and Ekman numbers. Elucidating the heat transport contributions of wall modes and of the boundary zonal flow are critical for characterizing the properties of the geostrophic regime of rotating convection in finite, physical containers and is crucial for connecting the geostrophic regime of laboratory convection with geophysical and astrophysical systems.

physics.flu-dyn↗

arXiv2021

LeViT-UNet: Make Faster Encoders with Transformer for Medical Image Segmentation

Medical image segmentation plays an essential role in developing computer-assisted diagnosis and therapy systems, yet still faces many challenges. In the past few years, the popular encoder-decoder architectures based on CNNs (e.g., U-Net) have been successfully applied in the task of medical image segmentation. However, due to the locality of convolution operations, they demonstrate limitations in learning global context and long-range spatial relations. Recently, several researchers try to introduce transformers to both the encoder and decoder components with promising results, but the efficiency requires further improvement due to the high computational complexity of transformers. In this paper, we propose LeViT-UNet, which integrates a LeViT Transformer module into the U-Net architecture, for fast and accurate medical image segmentation. Specifically, we use LeViT as the encoder of the LeViT-UNet, which better trades off the accuracy and efficiency of the Transformer block. Moreover, multi-scale feature maps from transformer blocks and convolutional blocks of LeViT are passed into the decoder via skip-connection, which can effectively reuse the spatial information of the feature maps. Our experiments indicate that the proposed LeViT-UNet achieves better performance comparing to various competing methods on several challenging medical image segmentation benchmarks including Synapse and ACDC. Code and models will be publicly available at https://github.com/apple1986/LeViT_UNet.

cs.CV↗

arXiv2021

Fast Quantum Property Prediction via Deeper 2D and 3D Graph Networks

Molecular property prediction is gaining increasing attention due to its diverse applications. One task of particular interests and importance is to predict quantum chemical properties without 3D equilibrium structures. This is practically favorable since obtaining 3D equilibrium structures requires extremely expensive calculations. In this work, we design a deep graph neural network to predict quantum properties by directly learning from 2D molecular graphs. In addition, we propose a 3D graph neural network to learn from low-cost conformer sets, which can be obtained with open-source tools using an affordable budget. We employ our methods to participate in the 2021 KDD Cup on OGB Large-Scale Challenge (OGB-LSC), which aims to predict the HOMO-LUMO energy gap of molecules. Final evaluation results reveal that we are one of the winners with a mean absolute error of 0.1235 on the holdout test set. Our implementation is available as part of the MoleculeX package (https://github.com/divelab/MoleculeX).

cs.LG↗

arXiv2021

An Integration and Operation Framework of Geothermal Heat Pumps in Distribution Networks

The application of the energy-efficient thermal and energy management mechanism in Geothermal Heat Pumps (GHPs) is indispensable to reduce the overall energy consumption and carbon emission across the building sector. Besides, in the Virtual Power Plant (VPP) system, the demand response of clustered GHP systems can improve the operating flexibility of the power grid. This paper presents an integration and operation framework of GHPs in the distribution network, applying a layered communication and optimization method to coordinate multiple clustered GHPs in a community. In the proposed hierarchical operation scheme, the operator of regional GHPs collects the thermal zone information and the disturbance prediction of buildings in a short time granularity, predicts the energy demand, and transmits the information to an aggregator. Using a novel linearized optimal power flow model, the aggregator coordinates and aggregates load resources of GHP systems in the distribution network. In this way, GHP systems with thermal and energy management mechanisms can be applied to achieve the demand response in the VPP and offer more energy flexibility to the community.

eess.SY↗

arXiv2021

Finding Physical Adversarial Examples for Autonomous Driving with Fast and Differentiable Image Compositing

There is considerable evidence that deep neural networks are vulnerable to adversarial perturbations applied directly to their digital inputs. However, it remains an open question whether this translates to vulnerabilities in real systems. For example, an attack on self-driving cars would in practice entail modifying the driving environment, which then impacts the video inputs to the car's controller, thereby indirectly leading to incorrect driving decisions. Such attacks require accounting for system dynamics and tracking viewpoint changes. We propose a scalable approach for finding adversarial modifications of a simulated autonomous driving environment using a differentiable approximation for the mapping from environmental modifications (rectangles on the road) to the corresponding video inputs to the controller neural network. Given the parameters of the rectangles, our proposed differentiable mapping composites them onto pre-recorded video streams of the original environment, accounting for geometric and color variations. Moreover, we propose a multiple trajectory sampling approach that enables our attacks to be robust to a car's self-correcting behavior. When combined with a neural network-based controller, our approach allows the design of adversarial modifications through end-to-end gradient-based optimization. Using the Carla autonomous driving simulator, we show that our approach is significantly more scalable and far more effective at identifying autonomous vehicle vulnerabilities in simulation experiments than a state-of-the-art approach based on Bayesian Optimization.

cs.CV↗

arXiv2021

Affinely representable lattices, stable matchings, and choice functions

Birkhoff's representation theorem (Birkhoff, 1937) defines a bijection between elements of a distributive lattice and the family of upper sets of an associated poset. Although not used explicitly, this result is at the backbone of the combinatorial algorithm by Irving et al. (1987) for maximizing a linear function over the set of stable matchings in Gale and Shapley's stable marriage model (Gale and Shapley, 1962). In this paper, we introduce a property of distributive lattices, which we term as affine representability, and show its role in efficiently solving linear optimization problems over the elements of a distributive lattice, as well as describing the convex hull of the characteristic vectors of the lattice elements. We apply this concept to the stable matching model with path-independent quota-filling choice functions, thus giving efficient algorithms and a compact polyhedral description for this model. To the best of our knowledge, this model generalizes all models from the literature for which similar results were known, and our paper is the first that proposes efficient algorithms for stable matchings with choice functions, beyond classical extensions of the Deferred Acceptance algorithm.

math.CO↗

arXiv2021

Sent2Matrix: Folding Character Sequences in Serpentine Manifolds for Two-Dimensional Sentence

We study text representation methods using deep models. Current methods, such as word-level embedding and character-level embedding schemes, treat texts as either a sequence of atomic words or a sequence of characters. These methods either ignore word morphologies or word boundaries. To overcome these limitations, we propose to convert texts into 2-D representations and develop the Sent2Matrix method. Our method allows for the explicit incorporation of both word morphologies and boundaries. When coupled with a novel serpentine padding method, our Sent2Matrix method leads to an interesting visualization in which 1-D character sequences are folded into 2-D serpentine manifolds. Notably, our method is the first attempt to represent texts in 2-D formats. Experimental results on text classification tasks shown that our method consistently outperforms prior embedding methods.

cs.CL↗

arXiv2021

On the law of the iterated logarithm for continued fractions with sequentially restricted partial quotients

We establish a law of the iterated logarithm (LIL) for the set of real numbers whose $n$-th partial quotient is bigger than $α_n$, where $(α_n)$ is a sequence such that $\sum 1/α_n$ is finite. This set is shown to have Hausdorff dimension $1/2$ in many cases and the measure in LIL is absolutely continuous to the Hausdorff measure. The result is obtained as an application of a strong invariance principle for unbounded observables on the limit set of a sequential iterated function system.

math.DS↗

arXiv2021

Boundary zonal flows in rapidly rotating turbulent thermal convection

Recently, in Zhang et al. (2020), it was found that in rapidly rotating turbulent Rayleigh-Bénard convection (RBC) in slender cylindrical containers (with diameter-to-height aspect ratio $Γ=1/2$) filled with a small-Prandtl-number fluid ($Pr \approx0.8$), the Large Scale Circulation (LSC) is suppressed and a Boundary Zonal Flow (BZF) develops near the sidewall, characterized by a bimodal PDF of the temperature, cyclonic fluid motion, and anticyclonic drift of the flow pattern (with respect to the rotating frame). This BZF carries a disproportionate amount ($>60\%$) of the total heat transport for $Pr < 1$ but decreases rather abruptly for larger $Pr$ to about $35\%$. In this work, we show that the BZF is robust and appears in rapidly rotating turbulent RBC in containers of different $Γ$ and in a broad range of $Pr$ and $Ra$. Direct numerical simulations for $0.1 \leq Pr \leq 12.3$, $10^7 \leq Ra \leq 5\times10^{9}$, $10^{5} \leq 1/Ek \leq 10^{7}$ and $Γ$ = 1/3, 1/2, 3/4, 1 and 2 show that the BZF width $δ_0$ scales with the Rayleigh number $Ra$ and Ekman number $Ek$ as $δ_0/H \sim Γ^{0} \Pr^{\{-1/4, 0\}} Ra^{1/4} Ek^{2/3}$ (${Pr<1, Pr>1}$) and the drift frequency as $ω/Ω\sim Γ^{0} Pr^{-4/3} Ra Ek^{5/3}$, where $H$ is the cell height and $Ω$ the angular rotation rate. The mode number of the BZF is 1 for $Γ\lesssim 1$ and $2 Γ$ for $Γ$ = {1,2} independent of $Ra$ and $Pr$. The BZF is quite reminiscent of wall mode states in rotating convection.

physics.flu-dyn↗