arXiv subjects

Xiao Xu

Publications and source records attributed to Xiao Xu.

At least 91 records · Page 5Linked to original sources

arXiv2023

Tractography-Based Parcellation of Cerebellar Dentate Nuclei via a Deep Nonnegative Matrix Factorization Clustering Method

As the largest human cerebellar nucleus, the dentate nucleus (DN) functions significantly in the communication between the cerebellum and the rest of the brain. Structural connectivity-based parcellation has the potential to reveal the topography of the DN and enable the study of its subregions. In this paper, we investigate a deep nonnegative matrix factorization clustering method (DNMFC) for parcellation of the human DN based on its structural connectivity using diffusion MRI tractography. We propose to describe the connectivity of the DN using a set of curated tractography fiber clusters within the cerebellum. Experiments are conducted on the diffusion MRI data of 50 healthy adults from the Human Connectome Project. In comparison with state-of-the-art clustering methods, DN parcellations resulting from DNMFC show better quality and consistency of parcels across subjects.

eess.IV↗

arXiv2023

Towards a unified nonlocal, peridynamics framework for the coarse-graining of molecular dynamics data with fractures

Molecular dynamics (MD) has served as a powerful tool for designing materials with reduced reliance on laboratory testing. However, the use of MD directly to treat the deformation and failure of materials at the mesoscale is still largely beyond reach. Herein, we propose a learning framework to extract a peridynamic model as a mesoscale continuum surrogate from MD simulated material fracture datasets. Firstly, we develop a novel coarse-graining method, to automatically handle the material fracture and its corresponding discontinuities in MD displacement dataset. Inspired by the Weighted Essentially Non-Oscillatory scheme, the key idea lies at an adaptive procedure to automatically choose the locally smoothest stencil, then reconstruct the coarse-grained material displacement field as piecewise smooth solutions containing discontinuities. Then, based on the coarse-grained MD data, a two-phase optimization-based learning approach is proposed to infer the optimal peridynamics model with damage criterion. In the first phase, we identify the optimal nonlocal kernel function from datasets without material damage, to capture the material stiffness properties. Then, in the second phase, the material damage criterion is learnt as a smoothed step function from the data with fractures. As a result, a peridynamics surrogate is obtained. Our peridynamics surrogate model can be employed in further prediction tasks with different grid resolutions from training, and hence allows for substantial reductions in computational cost compared with MD. We illustrate the efficacy of the proposed approach with several numerical tests for single layer graphene. Our tests show that the proposed data-driven model is robust and generalizable: it is capable in modeling the initialization and growth of fractures under discretization and loading settings that are different from the ones used during training.

cond-mat.mtrl-sci↗

arXiv2022

A precisely regulating phase evolution strategy for highly efficient kesterite solar cells

Phase evolution during the selenization is crucial for high-quality kesterite Cu2ZnSn(S, Se)4 (CZTSSe) absorbers and efficient solar cells. Herein, we regulate kinetic process of phase evolution from Cu+-Sn4+-MOE (MOE: 2-methoxyethanol) system by precisely controlling positive chamber pressure. We found that, at the heating-up stage, Se vapor concentration is intentionally suppressed in low-temperature region, which effectively reduces collision probability between the CZTS and Se atoms, thus remarkably inhibiting formation of secondary phases on the surface and multiple-step phase evolution processes. This strategy enables the phase evolution to start at relatively higher temperature and thereby leading to high crystalline quality CZTSSe absorber with fewer defects, and corresponding CZTSSe solar cell can present 14.1% efficiency (total area), which is the highest result so far. This work provides important insights into selenization mechanism of CZTSSe absorbers and explores a new way of kinetic regulation strategy to simplify the phase evolution path to efficient CZTSSe solar cells.

cond-mat.mtrl-sci↗

arXiv2022

Machine-learning of nonlocal kernels for anomalous subsurface transport from breakthrough curves

Anomalous behavior is ubiquitous in subsurface solute transport due to the presence of high degrees of heterogeneity at different scales in the media. Although fractional models have been extensively used to describe the anomalous transport in various subsurface applications, their application is hindered by computational challenges. Simpler nonlocal models characterized by integrable kernels and finite interaction length represent a computationally feasible alternative to fractional models; yet, the informed choice of their kernel functions still remains an open problem. We propose a general data-driven framework for the discovery of optimal kernels on the basis of very small and sparse data sets in the context of anomalous subsurface transport. Using spatially sparse breakthrough curves recovered from fine-scale particle-density simulations, we learn the best coarse-scale nonlocal model using a nonlocal operator regression technique. Predictions of the breakthrough curves obtained using the optimal nonlocal model show good agreement with fine-scale simulation results even at locations and time intervals different from the ones used to train the kernel, confirming the excellent generalization properties of the proposed algorithm. A comparison with trained classical models and with black-box deep neural networks confirms the superiority of the predictive capability of the proposed model.

math.NA↗

arXiv2022

Text is no more Enough! A Benchmark for Profile-based Spoken Language Understanding

Current researches on spoken language understanding (SLU) heavily are limited to a simple setting: the plain text-based SLU that takes the user utterance as input and generates its corresponding semantic frames (e.g., intent and slots). Unfortunately, such a simple setting may fail to work in complex real-world scenarios when an utterance is semantically ambiguous, which cannot be achieved by the text-based SLU models. In this paper, we first introduce a new and important task, Profile-based Spoken Language Understanding (ProSLU), which requires the model that not only relies on the plain text but also the supporting profile information to predict the correct intents and slots. To this end, we further introduce a large-scale human-annotated Chinese dataset with over 5K utterances and their corresponding supporting profile information (Knowledge Graph (KG), User Profile (UP), Context Awareness (CA)). In addition, we evaluate several state-of-the-art baseline models and explore a multi-level knowledge adapter to effectively incorporate profile information. Experimental results reveal that all existing text-based SLU models fail to work when the utterances are semantically ambiguous and our proposed framework can effectively fuse the supporting information for sentence-level intent detection and token-level slot filling. Finally, we summarize key challenges and provide new points for future directions, which hopes to facilitate the research.

cs.CL↗

arXiv2021

Don't be Contradicted with Anything! CI-ToD: Towards Benchmarking Consistency for Task-oriented Dialogue System

Consistency Identification has obtained remarkable success on open-domain dialogue, which can be used for preventing inconsistent response generation. However, in contrast to the rapid development in open-domain dialogue, few efforts have been made to the task-oriented dialogue direction. In this paper, we argue that consistency problem is more urgent in task-oriented domain. To facilitate the research, we introduce CI-ToD, a novel dataset for Consistency Identification in Task-oriented Dialog system. In addition, we not only annotate the single label to enable the model to judge whether the system response is contradictory, but also provide more fine-grained labels (i.e., Dialogue History Inconsistency, User Query Inconsistency and Knowledge Base Inconsistency) to encourage model to know what inconsistent sources lead to it. Empirical results show that state-of-the-art methods only achieve 51.3%, which is far behind the human performance of 93.2%, indicating that there is ample room for improving consistency identification ability. Finally, we conduct exhaustive experiments and qualitative analysis to comprehend key challenges and provide guidance for future directions. All datasets and models are publicly available at \url{https://github.com/yizhen20133868/CI-ToD}.

cs.CL↗

arXiv2021

Gibbs Paradox in the View of Information Entropy

This paper introduces the basic concepts of information theory. Based on these concepts, we regard the states in the state space and the types of ideal gases as the symbols in a symbol set to calculate the mixing entropy of ideal gas involved in Gibbs Paradox. The discussion above reveals that the non-need for distinguishing can resolve the contradiction of Gibbs Paradox, implying the introduction of indistinguishability is not necessary. Further analysis shows that the information entropy of gas molecular types does not directly correlate to the energy of a gas system, so it should not be used for calculating thermodynamic and statistical dynamic entropies. Therefore, the mixing entropy of the ideal gas is independent of the molecular types and is much smaller than the value commonly thought.

cond-mat.stat-mech↗

arXiv2021

Asset Selection via Correlation Blockmodel Clustering

We aim to cluster financial assets in order to identify a small set of stocks to approximate the level of diversification of the whole universe of stocks. We develop a data-driven approach to clustering based on a correlation blockmodel in which assets in the same cluster are highly correlated with each other and, at the same time, have the same correlations with all other assets. We devise an algorithm to detect the clusters, with theoretical analysis and practical guidance. Finally, we conduct an empirical analysis to verify the performance of the algorithm.

q-fin.PM↗

arXiv2021

GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling

Multi-intent SLU can handle multiple intents in an utterance, which has attracted increasing attention. However, the state-of-the-art joint models heavily rely on autoregressive approaches, resulting in two issues: slow inference speed and information leakage. In this paper, we explore a non-autoregressive model for joint multiple intent detection and slot filling, achieving more fast and accurate. Specifically, we propose a Global-Locally Graph Interaction Network (GL-GIN) where a local slot-aware graph interaction layer is proposed to model slot dependency for alleviating uncoordinated slots problem while a global intent-slot graph interaction layer is introduced to model the interaction between multiple intents and all slots in the utterance. Experimental results on two public datasets show that our framework achieves state-of-the-art performance while being 11.5 times faster.

cs.CL↗

arXiv2021

Deep learning for prediction of hepatocellular carcinoma recurrence after resection or liver transplantation: a discovery and validation study

This study aimed to develop a classifier of prognosis after resection or liver transplantation (LT) for HCC by directly analysing the ubiquitously available histological images using deep learning based neural networks. Nucleus map set was used to train U-net to capture the nuclear architectural information. Train set included the patients with HCC treated by resection and has a distinct outcome. LT set contained patients with HCC treated by LT. Train set and its nuclear architectural information extracted by U-net were used to train MobileNet V2 based classifier (MobileNetV2_HCC_Class), purpose-built for classifying supersized heterogeneous images. The MobileNetV2_HCC_Class maintained relative higher discriminatory power than the other factors after HCC resection or LT in the independent validation set. Pathological review showed that the tumoral areas most predictive of recurrence were characterized by presence of stroma, high degree of cytological atypia, nuclear hyperchomasia, and a lack of immune infiltration. A clinically useful prognostic classifier was developed using deep learning allied to histological slides. The classifier has been extensively evaluated in independent patient populations with different treatment, and gives consistent excellent results across the classical clinical, biological and pathological features. The classifier assists in refining the prognostic prediction of HCC patients and identifying patients who would benefit from more intensive management.

cs.CV↗

arXiv2021

A FETI approach to domain decomposition for meshfree discretizations of nonlocal problems

We propose a domain decomposition method for the efficient simulation of nonlocal problems. Our approach is based on a multi-domain formulation of a nonlocal diffusion problem where the subdomains share "nonlocal" interfaces of the size of the nonlocal horizon. This system of nonlocal equations is first rewritten in terms of minimization of a nonlocal energy, then discretized with a meshfree approximation and finally solved via a Lagrange multiplier approach in a way that resembles the finite element tearing and interconnect method. Specifically, we propose a distributed projected gradient algorithm for the solution of the Lagrange multiplier system, whose unknowns determine the nonlocal interface conditions between subdomains. Several two-dimensional numerical tests illustrate the strong and weak scalability of our algorithm, which outperforms the standard approach to the distributed numerical solution of the problem. This work is the first rigorous numerical study in a two-dimensional multi-domain setting for nonlocal operators with finite horizon and, as such, it is a fundamental step towards increasing the use of nonlocal models in large scale simulations.

math.NA↗

arXiv2021

Memory-Constrained No-Regret Learning in Adversarial Bandits

An adversarial bandit problem with memory constraints is studied where only the statistics of a subset of arms can be stored. A hierarchical learning policy that requires only a sublinear order of memory space in terms of the number of arms is developed. Its sublinear regret orders with respect to the time horizon are established for both weak regret and shifting regret. This work appears to be the first on memory-constrained bandit problems under the adversarial setting.

cs.LG↗

arXiv2021

A machine-learning framework for peridynamic material models with physical constraints

As a nonlocal extension of continuum mechanics, peridynamics has been widely and effectively applied in different fields where discontinuities in the field variables arise from an initially continuous body. An important component of the constitutive model in peridynamics is the influence function which weights the contribution of all the interactions over a nonlocal region surrounding a point of interest. Recent work has shown that in solid mechanics the influence function has a strong relationship with the heterogeneity of a material's micro-structure. However, determining an accurate influence function analytically from a given micro-structure typically requires lengthy derivations and complex mathematical models. To avoid these complexities, the goal of this paper is to develop a data-driven regression algorithm to find the optimal bond-based peridynamic model to describe the macro-scale deformation of linear elastic medium with periodic heterogeneity. We generate macro-scale deformation training data by averaging over periodic micro-structure unit cells and add a physical energy constraint representing the homogenized elastic modulus of the micro-structure to the regression algorithm. We demonstrate this scheme for examples of one- and two-dimensional linear elastodynamics and show that the energy constraint improves the accuracy of the resulting peridynamic model.

math.NA↗

arXiv2020

AGIF: An Adaptive Graph-Interactive Framework for Joint Multiple Intent Detection and Slot Filling

In real-world scenarios, users usually have multiple intents in the same utterance. Unfortunately, most spoken language understanding (SLU) models either mainly focused on the single intent scenario, or simply incorporated an overall intent context vector for all tokens, ignoring the fine-grained multiple intents information integration for token-level slot prediction. In this paper, we propose an Adaptive Graph-Interactive Framework (AGIF) for joint multiple intent detection and slot filling, where we introduce an intent-slot graph interaction layer to model the strong correlation between the slot and intents. Such an interaction layer is applied to each token adaptively, which has the advantage to automatically extract the relevant intents information, making a fine-grained intent information integration for the token-level slot prediction. Experimental results on three multi-intent datasets show that our framework obtains substantial improvement and achieves the state-of-the-art performance. In addition, our framework achieves new state-of-the-art performance on two single-intent datasets.

cs.CL↗

arXiv2020

Charge and hydration structure of dendritic polyelectrolytes: molecular simulations of polyglycerol sulphate

Macromolecules based on dendritic or hyperbranched polyelectrolytes have been emerging as high potential candidates for biomedical applications. Here we study the charge and solvation structure of dendritic polyglycerol sulphate (dPGS) of generations 0 to 3 in aqueous sodium chloride solution by explicit-solvent molecular dynamics computer simulations. We characterize dPGS by calculating several important properties such as relevant dPGS radii, molecular distributions, the solvent accessible surface area, and the partial molecular volume. In particular, as the dPGS exhibits high charge renormalization effects, we address the challenges of how to obtain a well-defined effective charge and surface potential of dPGS for practical applications. We compare implicit- and explicit-solvent approaches in our all-atom simulations with the coarse-grained simulations from our previous work. We find consistent values for the effective electrostatic size (i.e., the location of the effective charge of a Debye--Hückel sphere) within all the approaches, deviating at most by the size of a water molecule. Finally, the excess chemical potential of water insertion into dPGS and its thermodynamic signature are presented and rationalized.

cond-mat.soft↗

arXiv2020

Competitive sorption of mono- versus di-valent ions by highly charged globular macromolecules

When a highly charged globular macromolecule, such as a dendritic polyelectrolyte or charged nanogel, is immersed into a physiological electrolyte solution, monovalent and divalent counterions from the solution bind to the macromolecule in a certain ratio and thereby almost completely electroneutralize it. For charged macromolecules in biological media, the number ratio of bound mono- versus di-valent ions is decisive for the desired function. A theoretical prediction of such a sorption ratio is challenging because of the competition of electrostatic (valency), ion-specific, and binding saturation effects. Here, we devise and discuss a few approximate models to predict such an equilibrium sorption ratio by extending and combining established electrostatic binding theories such as Donnan, Langmuir, Manning as well as Poisson-Boltzmann approaches, to systematically study the competitive uptake of mono- and di-valent counterions by the macromolecule. We compare and fit our models to coarse-grained (implicit-solvent) computer simulation data of the globular polyelectrolyte dendritic polyglycerol sulfate (dPGS) in salt solutions of mixed valencies. The dPGS has high potential to serve in macromolecular carrier applications in biological systems and at the same time constitutes a good model system for a highly charged macromolecule. We finally use the simulation-informed models to extrapolate and predict electrostatic features such as the effective charge as a function of the divalent ion concentration for a wide range of dPGS generations (sizes).

cond-mat.soft↗

arXiv2020

Dynamic Fusion Network for Multi-Domain End-to-end Task-Oriented Dialog

Recent studies have shown remarkable success in end-to-end task-oriented dialog system. However, most neural models rely on large training data, which are only available for a certain number of task domains, such as navigation and scheduling. This makes it difficult to scalable for a new domain with limited labeled data. However, there has been relatively little research on how to effectively use data from all domains to improve the performance of each domain and also unseen domains. To this end, we investigate methods that can make explicit use of domain knowledge and introduce a shared-private network to learn shared and specific knowledge. In addition, we propose a novel Dynamic Fusion Network (DF-Net) which automatically exploit the relevance between the target domain and each domain. Results show that our model outperforms existing methods on multi-domain dialogue, giving the state-of-the-art in the literature. Besides, with little training data, we show its transferability by outperforming prior best model by 13.9\% on average.

cs.CL↗

arXiv2020

The optimal investment strategy of a DC pension plan under deposit loan spread and the O-U process

This paper is devoted to invest an optimal investment strategy for a defined-contribution (DC) pension plan under the Ornstein-Uhlenbeck (O-U) process and the loan. By considering risk-free asset, a risky asset driven by O-U process and a loan in the financial market, we firstly set up the dynamic equation and the asset market model which are instrumental in achieving the expected utility of ultimate wealth at retirement. Secondly, the corresponding Hamilton-Jacobi-Bellman(HJB) equation is derived by means of dynamic programming principle. The explicit expression for the optimal investment strategy is obtained by Legendre transform method. Finally, different parameters are selected to simulate the explicit solution and the financial interpretation of the optimal investment strategy is given.

q-fin.PM↗