arXiv subjects

Ang Li

Publications and source records attributed to Ang Li.

At least 397 records · Page 22Linked to original sources

arXiv2022

Probabilities of Causation: Adequate Size of Experimental and Observational Samples

The probabilities of causation are commonly used to solve decision-making problems. Tian and Pearl derived sharp bounds for the probability of necessity and sufficiency (PNS), the probability of sufficiency (PS), and the probability of necessity (PN) using experimental and observational data. The assumption is that one is in possession of a large enough sample to permit an accurate estimation of the experimental and observational distributions. In this study, we present a method for determining the sample size needed for such estimation, when a given confidence interval (CI) is specified. We further show by simulation that the proposed sample size delivered stable estimations of the bounds of PNS.

cs.AI↗

arXiv2022

Unit Selection: Case Study and Comparison with A/B Test Heuristic

The unit selection problem defined by Li and Pearl identifies individuals who have desired counterfactual behavior patterns, for example, individuals who would respond positively if encouraged and would not otherwise. Li and Pearl showed by example that their unit selection model is beyond the A/B test heuristics. In this paper, we reveal the essence of the A/B test heuristics, which are exceptional cases of the benefit function defined by Li and Pearl. Furthermore, We provided more simulated use cases of Li-Pearl's unit selection model to help decision-makers apply their model correctly, explaining that A/B test heuristics are generally problematic.

cs.AI↗

arXiv2022

Scientific Paper Classification Based on Graph Neural Network with Hypergraph Self-attention Mechanism

The number of scientific papers has increased rapidly in recent years. How to make good use of scientific papers for research is very important. Through the high-quality classification of scientific papers, researchers can quickly find the resource content they need from the massive scientific resources. The classification of scientific papers will effectively help researchers filter redundant information, obtain search results quickly and accurately, and improve the search quality, which is necessary for scientific resource management. This paper proposed a science-technique paper classification method based on hypergraph neural network(SPHNN). In the heterogeneous information network of scientific papers, the repeated high-order subgraphs are modeled as hyperedges composed of multiple related nodes. Then the whole heterogeneous information network is transformed into a hypergraph composed of different hyperedges. The graph convolution operation is carried out on the hypergraph structure, and the hyperedges self-attention mechanism is introduced to aggregate different types of nodes in the hypergraph, so that the final node representation can effectively maintain high-order nearest neighbor relationships and complex semantic information. Finally, by comparing with other methods, we proved that the model proposed in this paper has improved its performance.

cs.IR↗

arXiv2022

Embedding Representation of Academic Heterogeneous Information Networks Based on Federated Learning

Academic networks in the real world can usually be portrayed as heterogeneous information networks (HINs) with multi-type, universally connected nodes and multi-relationships. Some existing studies for the representation learning of homogeneous information networks cannot be applicable to heterogeneous information networks because of the lack of ability to issue heterogeneity. At the same time, data has become a factor of production, playing an increasingly important role. Due to the closeness and blocking of businesses among different enterprises, there is a serious phenomenon of data islands. To solve the above challenges, aiming at the data information of scientific research teams closely related to science and technology, we proposed an academic heterogeneous information network embedding representation learning method based on federated learning (FedAHE), which utilizes node attention and meta path attention mechanism to learn low-dimensional, dense and real-valued vector representations while preserving the rich topological information and meta-path-based semantic information of nodes in network. Moreover, we combined federated learning with the representation learning of HINs composed of scientific research teams and put forward a federal training mechanism based on dynamic weighted aggregation of parameters (FedDWA) to optimize the node embeddings of HINs. Through sufficient experiments, the efficiency, accuracy and feasibility of our proposed framework are demonstrated.

cs.IR↗

arXiv2022

Scientific and Technological News Recommendation Based on Knowledge Graph with User Perception

Existing research usually utilizes side information such as social network or item attributes to improve the performance of collaborative filtering-based recommender systems. In this paper, the knowledge graph with user perception is used to acquire the source of side information. We proposed KGUPN to address the limitations of existing embedding-based and path-based knowledge graph-aware recommendation methods, an end-to-end framework that integrates knowledge graph and user awareness into scientific and technological news recommendation systems. KGUPN contains three main layers, which are the propagation representation layer, the contextual information layer and collaborative relation layer. The propagation representation layer improves the representation of an entity by recursively propagating embeddings from its neighbors (which can be users, news, or relationships) in the knowledge graph. The contextual information layer improves the representation of entities by encoding the behavioral information of entities appearing in the news. The collaborative relation layer complements the relationship between entities in the news knowledge graph. Experimental results on real-world datasets show that KGUPN significantly outperforms state-of-the-art baselines in scientific and technological news recommendation.

cs.IR↗

arXiv2022

Rethinking Normalization Methods in Federated Learning

Federated learning (FL) is a popular distributed learning framework that can reduce privacy risks by not explicitly sharing private data. In this work, we explicitly uncover external covariate shift problem in FL, which is caused by the independent local training processes on different devices. We demonstrate that external covariate shifts will lead to the obliteration of some devices' contributions to the global model. Further, we show that normalization layers are indispensable in FL since their inherited properties can alleviate the problem of obliterating some devices' contributions. However, recent works have shown that batch normalization, which is one of the standard components in many deep neural networks, will incur accuracy drop of the global model in FL. The essential reason for the failure of batch normalization in FL is poorly studied. We unveil that external covariate shift is the key reason why batch normalization is ineffective in FL. We also show that layer normalization is a better choice in FL which can mitigate the external covariate shift and improve the performance of the global model. We conduct experiments on CIFAR10 under non-IID settings. The results demonstrate that models with layer normalization converge fastest and achieve the best or comparable accuracy for three different model architectures.

cs.LG↗

arXiv2022

Towards Collaborative Intelligence: Routability Estimation based on Decentralized Private Data

Applying machine learning (ML) in design flow is a popular trend in EDA with various applications from design quality predictions to optimizations. Despite its promise, which has been demonstrated in both academic researches and industrial tools, its effectiveness largely hinges on the availability of a large amount of high-quality training data. In reality, EDA developers have very limited access to the latest design data, which is owned by design companies and mostly confidential. Although one can commission ML model training to a design company, the data of a single company might be still inadequate or biased, especially for small companies. Such data availability problem is becoming the limiting constraint on future growth of ML for chip design. In this work, we propose an Federated-Learning based approach for well-studied ML applications in EDA. Our approach allows an ML model to be collaboratively trained with data from multiple clients but without explicit access to the data for respecting their data privacy. To further strengthen the results, we co-design a customized ML model FLNet and its personalization under the decentralized training scenario. Experiments on a comprehensive dataset show that collaborative training improves accuracy by 11% compared with individual local models, and our customized model FLNet significantly outperforms the best of previous routability estimators in this collaborative training flow.

cs.LG↗

arXiv2022

Low Complexity SLP: An Inversion-Free, Parallelizable ADMM Approach

We propose a parallel constructive interference (CI)-based symbol-level precoding (SLP) approach for massive connectivity in the downlink of multiuser multiple-input single-output (MU-MISO) systems, with only local channel state information (CSI) used at each processor unit and limited information exchange between processor units. By reformulating the power minimization (PM) SLP problem and exploiting the separability of the corresponding reformulation, the original problem is decomposed into several parallel subproblems via the ADMM framework with closed-form solutions, leading to a substantial reduction in computational complexity. The sufficient condition for guaranteeing the convergence of the proposed approach is derived, based on which an adaptive parameter tuning strategy is proposed to accelerate the convergence rate. To avoid the large-dimension matrix inverse operation, an efficient algorithm is proposed by employing the standard proximal term and by leveraging the singular value decomposition (SVD). Furthermore, a prox-linear proximal term is adopted to fully eliminate the matrix inversion, and a parallel inverse-free SLP (PIF-SLP) algorithm is finally obtained. Numerical results validate our derivations above, and demonstrate that the proposed PIF-SLP algorithm can significantly reduce the computational complexity compared to the state-of-the-arts.

cs.IT↗

arXiv2022

QuGAN: A Quantum State Fidelity based Generative Adversarial Network

Tremendous progress has been witnessed in artificial intelligence where neural network backed deep learning systems have been used, with applications in almost every domain. As a representative deep learning framework, Generative Adversarial Network (GAN) has been widely used for generating artificial images, text-to-image or image augmentation across areas of science, arts and video games. However, GANs are computationally expensive, sometimes computationally prohibitive. Furthermore, training GANs may suffer from convergence failure and modal collapse. Aiming at the acceleration of use cases for practical quantum computers, we propose QuGAN, a quantum GAN architecture that provides stable convergence, quantum-state based gradients and significantly reduced parameter sets. The QuGAN architecture runs both the discriminator and the generator purely on quantum state fidelity and utilizes the swap test on qubits to calculate the values of quantum-based loss functions. Built on quantum layers, QuGAN achieves similar performance with a 94.98% reduction on the parameter set when compared to classical GANs. With the same number of parameters, additionally, QuGAN outperforms state-of-the-art quantum based GANs in the literature providing a 48.33% improvement in system performance compared to others attaining less than 0.5% in terms of similarity between generated distributions and original data sets. QuGAN code is released at https://github.com/yingmao/Quantum-Generative-Adversarial-Network

quant-ph↗

arXiv2022

MSREP: A Fast yet Light Sparse Matrix Framework for Multi-GPU Systems

Sparse linear algebra kernels play a critical role in numerous applications, covering from exascale scientific simulation to large-scale data analytics. Offloading linear algebra kernels on one GPU will no longer be viable in these applications, simply because the rapidly growing data volume may exceed the memory capacity and computing power of a single GPU. Multi-GPU systems nowadays being ubiquitous in supercomputers and data-centers present great potentials in scaling up large sparse linear algebra kernels. In this work, we design a novel sparse matrix representation framework for multi-GPU systems called MSREP, to scale sparse linear algebra operations based on our augmented sparse matrix formats in a balanced pattern. Different from dense operations, sparsity significantly intensifies the difficulty of distributing the computation workload among multiple GPUs in a balanced manner. We enhance three mainstream sparse data formats -- CSR, CSC, and COO, to enable fine-grained data distribution. We take sparse matrix-vector multiplication (SpMV) as an example to demonstrate the efficiency of our MSREP framework. In addition, MSREP can be easily extended to support other sparse linear algebra kernels based on the three fundamental formats (i.e., CSR, CSC and COO).

cs.DC↗

arXiv2022

CollComm: Enabling Efficient Collective Quantum Communication Based on EPR buffering

The noisy and lengthy nature of quantum communication hinders the development of distributed quantum computing. The inefficient design of existing compilers for distributed quantum computing worsens the situation. Previous compilation frameworks couple communication hardware with the implementation of expensive remote gates. However, we discover that the efficiency of quantum communication, especially collective communication, can be significantly boosted by decoupling communication resources from remote operations, that is, the communication hardware would be used only for preparing remote entanglement, and the computational hardware, the component used to store program information, would be used for conducting remote gates. Based on the observation, we develop a compiler framework to optimize the collective communication happening in distributed quantum programs. In this framework, we decouple the communication preparation process in communication hardware from the remote gates conducted in computational hardware by buffering EPR pairs generated by communication hardware in qubits of the computational hardware. Experimental results show that the proposed framework can halve the communication cost of various distributed quantum programs, compared to state-of-the-art compilers for distributed quantum computing.

quant-ph↗

arXiv2022

Unified neutron star EOSs and neutron star structures in RMF models

In the framework of Thomas-Fermi approximation, we study systematically the EOSs and microscopic structures of neutron star matter in a vast density range with $n_\mathrm{b}\approx 10^{-10}$-2 $\mathrm{fm}^{-3}$, where various covariant density functionals are adopted, i.e., those with nonlinear self couplings (NL3, PK1, TM1, GM1, MTVTC) and density-dependent couplings (DD-LZ1, DDME-X, PKDD, DD-ME2, DD2, TW99). It is found that the EOSs generally coincide with each other at $n_\mathrm{b}\lesssim 10^{-4}$ fm${}^{-3}$ and 0.1 fm${}^{-3}\lesssim n_\mathrm{b} \lesssim 0.3$ fm${}^{-3}$, while in other density regions they are sensitive to the effective interactions between nucleons. By adopting functionals with larger slope of symmetry energy $L$, the curvature parameter $K_\mathrm{sym}$ and neutron drip density generally increase, while the droplet size, proton number of nucleus, core-crust transition density, and onset density of non-spherical nuclei decrease. All functionals predict neutron stars with maximum masses exceeding the two-solar-mass limit, while those of DD2, DD-LZ1, DD-ME2, and DDME-X predict optimum neutron star radii according to the observational constraints. Nevertheless, the corresponding skewness coefficients $J$ are much lager than expected, while only the functionals MTVTC and TW99 meet the start-of-art constraints on $J$. More accurate measurements on the radius of PSR J0740+6620 and the maximum mass of neutron stars are thus essential to identify the functional that satisfies all constraints from nuclear physics and astrophysical observations. Approximate linear correlations between neutron stars' radii at $M=1.4 M_{\odot}$ and $2 M_{\odot}$, the slope $L$ and curvature parameter $K_\mathrm{sym}$ of symmetry energy are observed as well, which is mainly attributed to the curvature-slope correlations in the functionals adopted here.

nucl-th↗

arXiv2022

Unit Selection with Nonbinary Treatment and Effect

The unit selection problem aims to identify a set of individuals who are most likely to exhibit a desired mode of behavior, for example, selecting individuals who would respond one way if encouraged and a different way if not encouraged. Using a combination of experimental and observational data, Li and Pearl derived tight bounds on the "benefit function", which is the payoff/cost associated with selecting an individual with given characteristics. This paper extends the benefit function to the general form such that the treatment and effect are not restricted to binary. We propose an algorithm to test the identifiability of the nonbinary benefit function and an algorithm to compute the bounds of the nonbinary benefit function using experimental and observational data.

cs.AI↗

arXiv2022

A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining

Transformers are considered one of the most important deep learning models since 2018, in part because it establishes state-of-the-art (SOTA) records and could potentially replace existing Deep Neural Networks (DNNs). Despite the remarkable triumphs, the prolonged turnaround time of Transformer models is a widely recognized roadblock. The variety of sequence lengths imposes additional computing overhead where inputs need to be zero-padded to the maximum sentence length in the batch to accommodate the parallel computing platforms. This paper targets the field-programmable gate array (FPGA) and proposes a coherent sequence length adaptive algorithm-hardware co-design for Transformer acceleration. Particularly, we develop a hardware-friendly sparse attention operator and a length-aware hardware resource scheduling algorithm. The proposed sparse attention operator brings the complexity of attention-based models down to linear complexity and alleviates the off-chip memory traffic. The proposed length-aware resource hardware scheduling algorithm dynamically allocates the hardware resources to fill up the pipeline slots and eliminates bubbles for NLP tasks. Experiments show that our design has very small accuracy loss and has 80.2 $\times$ and 2.6 $\times$ speedup compared to CPU and GPU implementation, and 4 $\times$ higher energy efficiency than state-of-the-art GPU accelerator optimized via CUBLAS GEMM.

cs.LG↗

arXiv2022

Probabilities of Causation with Nonbinary Treatment and Effect

This paper deals with the problem of estimating the probabilities of causation when treatment and effect are not binary. Tian and Pearl derived sharp bounds for the probability of necessity and sufficiency (PNS), the probability of sufficiency (PS), and the probability of necessity (PN) using experimental and observational data. In this paper, we provide theoretical bounds for all types of probabilities of causation to multivalued treatments and effects. We further discuss examples where our bounds guide practical decisions and use simulation studies to evaluate how informative the bounds are for various combinations of data.

cs.AI↗

arXiv2022

On the moment of inertia of PSR J0737-3039 A from LIGO/Virgo and NICER

We perform a Bayesian analysis of neutrons star moment of inertia by utilizing the available gravitational-wave data from LIGO/Virgo (GW170817 and GW190425) and mass-radius measurements from the Neutron Star Interior Composition Explorer (PSR J0030+0415 and PSR J0740+6620), incorporating the possible phase transition in the pulsar inner core. We find that the moment of inertia of pulsar A in the double pulsar binary J0737-3039 is $\sim1.30\times10^{45}\,{\rm g\,cm^2}$, which only slightly depends on the employed hadronic equation of states. We also demonstrate how a moment of inertia measurement would improve our knowledge of the equation of state and the mass-radius relation for neutron stars and discuss whether a quark deconfinement phase transition is supported by the available data and forthcoming data that could be consistent with this hypothesis. We find that if pulsar A is a quark star, that its moment of inertia is a large value of $\sim1.55\times10^{45}\,{\rm g\,cm^2}$ suggesting the possibility of distinguishing it from (hybrid-)neutron stars with measurements of PSR J0737-3039A moment of inertia. We finally demonstrate the moment-of-inertia-compactness universal relations and provide analytical fits for both (hybrid-)neutron star and quark star results based on our analysis.

astro-ph.HE↗

arXiv2022

Dark matter admixed neutron star properties in the light of X-ray pulse profile observations

The distribution of the dark matter (DM) in DM-admixed-neutron stars (DANSs) is supposed to be either a dense dark core or an extended dark halo, which is subject to the DM fraction of DANS ($f_χ$) and the DM properties, such as the mass ($m_χ$) and the strength of the self-interaction ($y$). In this paper, we perform an in-depth analysis of the formation criterion for dark core/dark halo and point out that the relative distribution of these two components is essentially determined by the ratio of the central enthalpy of the DM component to that of the baryonic matter component inside DANSs. For the critical case where the radii of DM and baryonic matter are the same, we further derive an analytical formula to describe the dependence of $f^{\rm crit}_χ$ on $m_χ$ and $y$ for given DANS mass. The relative distribution of the two components in DANSs can lead to different observational effects. We here focus on the modification of the pulsar pulse profile due to the extra light-bending effect in the case of a dark-halo existence and conduct the first investigation of the dark-halo effects on the pulse profile. We find that the peak flux deviation is strongly dependent on the ratio of the halo mass to the radius of the DM component. Lastly, we perform Bayesian parameter estimation on the DM particle properties based on the recent X-ray observations of PSR J0030+0451 and PSR J0740+6620 by the Neutron Star Interior Composition Explorer.

astro-ph.HE↗

arXiv2022

SphereFed: Hyperspherical Federated Learning

Federated Learning aims at training a global model from multiple decentralized devices (i.e. clients) without exchanging their private local data. A key challenge is the handling of non-i.i.d. (independent identically distributed) data across multiple clients that may induce disparities of their local features. We introduce the Hyperspherical Federated Learning (SphereFed) framework to address the non-i.i.d. issue by constraining learned representations of data points to be on a unit hypersphere shared by clients. Specifically, all clients learn their local representations by minimizing the loss with respect to a fixed classifier whose weights span the unit hypersphere. After federated training in improving the global model, this classifier is further calibrated with a closed-form solution by minimizing a mean squared loss. We show that the calibration solution can be computed efficiently and distributedly without direct access of local data. Extensive experiments indicate that our SphereFed approach is able to improve the accuracy of multiple existing federated learning algorithms by a considerable margin (up to 6% on challenging datasets) with enhanced computation and communication efficiency across datasets and model architectures.

cs.LG↗