arXiv subjects

Han Xiao

Publications and source records attributed to Han Xiao.

At least 127 records · Page 7Linked to original sources

arXiv2022

Shapley-NAS: Discovering Operation Contribution for Neural Architecture Search

In this paper, we propose a Shapley value based method to evaluate operation contribution (Shapley-NAS) for neural architecture search. Differentiable architecture search (DARTS) acquires the optimal architectures by optimizing the architecture parameters with gradient descent, which significantly reduces the search cost. However, the magnitude of architecture parameters updated by gradient descent fails to reveal the actual operation importance to the task performance and therefore harms the effectiveness of obtained architectures. By contrast, we propose to evaluate the direct influence of operations on validation accuracy. To deal with the complex relationships between supernet components, we leverage Shapley value to quantify their marginal contributions by considering all possible combinations. Specifically, we iteratively optimize the supernet weights and update the architecture parameters by evaluating operation contributions via Shapley value, so that the optimal architectures are derived by selecting the operations that contribute significantly to the tasks. Since the exact computation of Shapley value is NP-hard, the Monte-Carlo sampling based algorithm with early truncation is employed for efficient approximation, and the momentum update mechanism is adopted to alleviate fluctuation of the sampling process. Extensive experiments on various datasets and various search spaces show that our Shapley-NAS outperforms the state-of-the-art methods by a considerable margin with light search cost. The code is available at https://github.com/Euphoria16/Shapley-NAS.git

cs.LG↗

arXiv2022

AI Enlightens Wireless Communication: A Transformer Backbone for CSI Feedback

This paper is based on the background of the 2nd Wireless Communication Artificial Intelligence (AI) Competition (WAIC) which is hosted by IMT-2020(5G) Promotion Group 5G+AIWork Group, where the framework of the eigenvector-based channel state information (CSI) feedback problem is firstly provided. Then a basic Transformer backbone for CSI feedback referred to EVCsiNet-T is proposed. Moreover, a series of potential enhancements for deep learning based (DL-based) CSI feedback including i) data augmentation, ii) loss function design, iii) training strategy, and iv) model ensemble are introduced. The experimental results involving the comparison between EVCsiNet-T and traditional codebook methods over different channels are further provided, which show the advanced performance and a promising prospect of Transformer on DL-based CSI feedback problem.

eess.SP↗

arXiv2022

Support Vector Machines under Adversarial Label Contamination

Machine learning algorithms are increasingly being applied in security-related tasks such as spam and malware detection, although their security properties against deliberate attacks have not yet been widely understood. Intelligent and adaptive attackers may indeed exploit specific vulnerabilities exposed by machine learning techniques to violate system security. Being robust to adversarial data manipulation is thus an important, additional requirement for machine learning algorithms to successfully operate in adversarial settings. In this work, we evaluate the security of Support Vector Machines (SVMs) to well-crafted, adversarial label noise attacks. In particular, we consider an attacker that aims to maximize the SVM's classification error by flipping a number of labels in the training data. We formalize a corresponding optimal attack strategy, and solve it by means of heuristic approaches to keep the computational complexity tractable. We report an extensive experimental analysis on the effectiveness of the considered attacks against linear and non-linear SVMs, both on synthetic and real-world datasets. We finally argue that our approach can also provide useful insights for developing more secure SVM learning algorithms, and also novel techniques in a number of related research areas, such as semi-supervised and active learning.

cs.LG↗

arXiv2022

Variation between Antiferromagnetism and Ferrimagnetism in NiPS3 by Electron Doping

How to electrically control magnetic properties of a magnetic material is promising towards spintronic applications, where the investigation of carrier doping effects on antiferromagnetic (AFM) materials remains challenging due to their zero net magnetization. In this work, we found electron doping dependent variation of magnetic orders of a two-dimensional (2D) AFM insulator NiPS3, where doping concentration is tuned by intercalating various organic cations into the van der Waals gaps of NiPS3 without introduction of defects and impurity phases. The doped NiPS3 shows an AFM-ferrimagnetic (FIM) transition at doping level of 0.2-0.5 electrons/cell and a FIM-AFM transition at doping level of >= 0.6 electrons/cell. We propose that the found phenomenon is due to competition between Stoner exchange dominated inter-chain ferromagnetic order and super-exchange dominated inter-chain AFM order at different doping level. Our studies provide a viable way to exploit correlation between electronic structures and magnetic properties of 2D magnetic materials for realization of magnetoelectric effect.

cond-mat.mtrl-sci↗

arXiv2022

Arboricity games: the core and the nucleolus

The arboricity of a graph is the minimum number of forests required to cover all its edges. In this paper, we examine arboricity from a game-theoretic perspective and investigate cost-sharing in the minimum forest cover problem. We introduce the arboricity game as a cooperative cost game defined on a graph. The players are edges, and the cost of each coalition is the arboricity of the subgraph induced by the coalition. We study properties of the core and propose an efficient algorithm for computing the nucleolus when the core is not empty. In order to compute the nucleolus in the core, we introduce the prime partition which is built on the densest subgraph lattice. The prime partition decomposes the edge set of a graph into a partially ordered set defined from minimal densest minors and their invariant precedence relation. Moreover, edges from the same partition always have the same value in a core allocation. Consequently, when the core is not empty, the prime partition significantly reduces the number of variables and constraints required in the linear programs of Maschler's scheme and allows us to compute the nucleolus in polynomial time. Besides, the prime partition provides a graph decomposition analogous to the celebrated core decomposition and the density-friendly decomposition, which may be of independent interest.

cs.GT↗

arXiv2021

Excite atom-photon bound state inside the coupled-resonator waveguide coupled with a giant atom

It is of fundamental interest in controlling the light-matter interaction for a long time in the field of quantum information processing. However, usual excitation with the propagating photon can hardly excite a localized state of light while keeping the atom under a subradiant decay in an atom-waveguide system. Here, we propose a model of coupling between a giant atom and the dynamically-modulated coupled-resonator waveguide and find that a bound state, where the light shows the localization effect and atom exhibits a subradiant decay time, can be excited by a propagating photon. An analytical treatment based on the separation of the propagating states and localized states of light has been used and provides inspiring explanation of our finding, i.e., a propagating photon can be efficiently converted to the localized light through the light-atom interactions in three resonators at frequency difference precisely equivalent to external modulation frequency. Our work therefore provides an alternative method for actively localizing the photon in a modulated coupled-resonator waveguide system interacting with giant atom, and also points out a way to study the light-atom interaction in a synthetic frequency dimension that holds the similar Hamiltonian.

quant-ph↗

arXiv2021

Approximate Core Allocations for Multiple Partners Matching Games

The matching game is a cooperative game where the value of every coalition is the maximum revenue of players in the coalition can make by forming pairwise disjoint partners. The multiple partners matching game generalizes the matching game by allowing each player to have more than one possibly repeated partner. In this paper, we study profit-sharing in multiple partners matching games. A central concept for profit-sharing is the core which consists of all possible ways of distributing the profit among individual players such that the grand coalition remains intact. The core of multiple partners matching games may be empty [Deng et al., Algorithmic aspects of the core of combinatorial optimization games, Math. Oper. Res., 1999.]; even when the core is non-empty, the core membership problem is intractable in general [Biro et al., The stable fixtures problem with payments, Games Econ. Behav., 2018]. Thus we study approximate core allocations upon which a coalition may be paid less than the profit it makes by seceding from the grand coalition. We provide an LP-based mechanism guaranteeing that no coalition is paid less than $2/3$ times the profit it makes on its own. We also show that $2/3$ is the best possible factor relative to the underlying LP-relaxation. Our result generalizes the work of Vazirani [Vazirani, The general graph matching game: approximate core, arXiv, 2021] from matching games to multiple partners matching games.

cs.GT↗

arXiv2021

Multi-linear Tensor Autoregressive Models

Contemporary time series analysis has seen more and more tensor type data, from many fields. For example, stocks can be grouped according to Size, Book-to-Market ratio, and Operating Profitability, leading to a 3-way tensor observation at each month. We propose an autoregressive model for the tensor-valued time series, with autoregressive terms depending on multi-linear coefficient matrices. Comparing with the traditional approach of vectoring the tensor observations and then applying the vector autoregressive model, the tensor autoregressive model preserves the tensor structure and admits corresponding interpretations. We introduce three estimators based on projection, least squares, and maximum likelihood. Our analysis considers both fixed dimensional and high dimensional settings. For the former we establish the central limit theorems of the estimators, and for the latter we focus on the convergence rates and the model selection. The performance of the model is demonstrated by simulated and real examples.

stat.ME↗

arXiv2021

Generalizable Mixed-Precision Quantization via Attribution Rank Preservation

In this paper, we propose a generalizable mixed-precision quantization (GMPQ) method for efficient inference. Conventional methods require the consistency of datasets for bitwidth search and model deployment to guarantee the policy optimality, leading to heavy search cost on challenging largescale datasets in realistic applications. On the contrary, our GMPQ searches the mixed-quantization policy that can be generalized to largescale datasets with only a small amount of data, so that the search cost is significantly reduced without performance degradation. Specifically, we observe that locating network attribution correctly is general ability for accurate visual analysis across different data distribution. Therefore, despite of pursuing higher model accuracy and complexity, we preserve attribution rank consistency between the quantized models and their full-precision counterparts via efficient capacity-aware attribution imitation for generalizable mixed-precision quantization strategy search. Extensive experiments show that our method obtains competitive accuracy-complexity trade-off compared with the state-of-the-art mixed-precision networks in significantly reduced search cost. The code is available at https://github.com/ZiweiWangTHU/GMPQ.git.

cs.CV↗

arXiv2021

AI Enlightens Wireless Communication: Analyses, Solutions and Opportunities on CSI Feedback

In this paper, we give a systematic description of the 1st Wireless Communication Artificial Intelligence (AI) Competition (WAIC) which is hosted by IMT-2020(5G) Promotion Group 5G+AI Work Group. Firstly, the framework of full channel state information (F-CSI) feedback problem and its corresponding channel dataset are provided. Then the enhancing schemes for DL-based F-CSI feedback including i) channel data analysis and preprocessing, ii) neural network design and iii) quantization enhancement are elaborated. The final competition results composed of different enhancing schemes are presented. Based on the valuable experience of 1st WAIC, we also list some challenges and potential study areas for the design of AI-based wireless communication systems.

eess.SP↗

arXiv2021

Generalized Autoregressive Moving Average Models with GARCH Errors

One of the important and widely used classes of models for non-Gaussian time series is the generalized autoregressive model average models (GARMA), which specifies an ARMA structure for the conditional mean process of the underlying time series. However, in many applications one often encounters conditional heteroskedasticity. In this paper we propose a new class of models, referred to as GARMA-GARCH models, that jointly specify both the conditional mean and conditional variance processes of a general non-Gaussian time series. Under the general modeling framework, we propose three specific models, as examples, for proportional time series, nonnegative time series, and skewed and heavy-tailed financial time series. Maximum likelihood estimator (MLE) and quasi Gaussian MLE (GMLE) are used to estimate the parameters. Simulation studies and three applications are used to demonstrate the properties of the models and the estimation procedures.

stat.ME↗

arXiv2021

Population Monotonicity in Matching Games

A matching game is a cooperative profit game defined on an edge-weighted graph, where the players are the vertices and the profit of a coalition is the maximum weight of matchings in the subgraph induced by the coalition. A population monotonic allocation scheme is a collection of rules defining how to share the profit among players in each coalition such that every player is better off when the coalition expands. In this paper, we study matching games and provide a necessary and sufficient characterization for the existence of population monotonic allocation schemes. Our characterization also implies that whether a matching game admits population monotonic allocation schemes can be determined efficiently.

cs.GT↗

arXiv2020

Matrix Completion using Kronecker Product Approximation

A matrix completion problem is to recover the missing entries in a partially observed matrix. Most of the existing matrix completion methods assume a low rank structure of the underlying complete matrix. In this paper, we introduce an alternative and more general form of the underlying complete matrix, which assumes a low Kronecker rank instead of a low regular rank, but includes the latter as a special case. The extra flexibility allows for a much more parsimonious representation of the underlying matrix, but also raises the challenge of determining the proper Kronecker product configuration to be used. We find that the configuration can be identified using the mean squared error criterion as well as a modified cross-validation criterion. We establish the consistency of this procedure under suitable conditions on the signal-to-noise ratio. A aggregation procedure is also proposed to deal with special missing patterns and complex underlying structures. Both numerical and empirical studies are carried out to demonstrate the performance of the new method.

stat.ML↗

arXiv2020

On the Convexity of Independent Set Games

Independent set games are cooperative games defined on graphs, where players are edges and the value of a coalition is the maximum cardinality of independent sets in the subgraph defined by the coalition. In this paper, we investigate the convexity of independent set games, as convex games possess many nice properties both economically and computationally. For independent set games introduced by Deng et al. (Math. Oper. Res., 24:751-766, 1999), we provide a necessary and sufficient characterization for the convexity, i.e., every non-pendant edge is incident to a pendant edge in the underlying graph. Our characterization immediately yields a polynomial time algorithm for recognizing convex instances of independent set games. Besides, we introduce a new class of independent set games and provide an efficient characterization for the convexity.

cs.DM↗

arXiv2020

KoPA: Automated Kronecker Product Approximation

We consider the problem of matrix approximation and denoising induced by the Kronecker product decomposition. Specifically, we propose to approximate a given matrix by the sum of a few Kronecker products of matrices, which we refer to as the Kronecker product approximation (KoPA). Because the Kronecker product is an extension of the outer product from vectors to matrices, KoPA extends the low rank matrix approximation, and includes it as a special case. Comparing with the latter, KoPA also offers a greater flexibility, since it allows the user to choose the configuration, which are the dimensions of the two smaller matrices forming the Kronecker product. On the other hand, the configuration to be used is usually unknown, and needs to be determined from the data in order to achieve the optimal balance between accuracy and parsimony. We propose to use extended information criteria to select the configuration. Under the paradigm of high dimensional analysis, we show that the proposed procedure is able to select the true configuration with probability tending to one, under suitable conditions on the signal-to-noise ratio. We demonstrate the superiority of KoPA over the low rank approximations through numerical studies, and several benchmark image examples.

math.ST↗

arXiv2020

Population Monotonic Allocation Schemes for Vertex Cover Games

For vertex cover games (introduced by Deng et al., Math. Oper. Res., 24:751-766, 1999), we investigate population monotonic allocation schemes (introduced by Sprumont, Games Econ. Behav., 2: 378-394, 1990). We show that the existence of a population monotonic allocation scheme (PMAS for short) for vertex cover games can be determined efficiently and that a PMAS, if exists, can be constructed accordingly. We also show that integral PMAS-es for vertex cover games can be characterized with stable matchings and be enumerated by employing Gale-Shapley algorithm (introduced by Gale and Shapley, Amer. Math. Monthly, 69:9-15, 1962).

cs.GT↗

arXiv2020

COVID-19 societal response captured by seismic noise in China and Italy

Seismic noise with frequencies above 1 Hz is often called cultural noise and is generally correlated quite well with human activities. Recently, cities in mainland China and Italy imposed lockdown restrictions in response to COVID-19, which gave us an unprecedented opportunity to study the relationship between seismic noise above 1 Hz and human activities. Using seismic records from stations in China and Italy, we show that seismic noise above 1 Hz was primarily generated by the local transportation systems. The lockdown of the cities and the imposition of travel restrictions led to a ~4-12 dB energy decrease in seismic noise in mainland China. Data also show that different Chinese cities experienced distinct periods of diminished cultural noise, related to differences in local response to the epidemic. In contrast, there was only ~1-6 dB energy decrease of seismic noise in Italy, after the country was put under a lockdown. The noise data indicate that traffic flow did not decrease as much in Italy, but show how different cities reacted distinctly to the lockdown conditions.

physics.geo-ph↗

arXiv2020

KSR: A Semantic Representation of Knowledge Graph within a Novel Unsupervised Paradigm

Knowledge representation is a long-history topic in AI, which is very important. A variety of models have been proposed for knowledge graph embedding, which projects symbolic entities and relations into continuous vector space. However, most related methods merely focus on the data-fitting of knowledge graph, and ignore the interpretable semantic expression. Thus, traditional embedding methods are not friendly for applications that require semantic analysis, such as question answering and entity retrieval. To this end, this paper proposes a semantic representation method for knowledge graph \textbf{(KSR)}, which imposes a two-level hierarchical generative process that globally extracts many aspects and then locally assigns a specific category in each aspect for every triple. Since both aspects and categories are semantics-relevant, the collection of categories in each aspect is treated as the semantic representation of this triple. Extensive experiments show that our model outperforms other state-of-the-art baselines substantially.

cs.LG↗