arXiv subjects
Publications and source records attributed to Jian-Guo Liu.
We prove local existence of smooth solutions for large data and global smooth solutions for small data to the incompressible, resitive, viscous or inviscid Hall-MHD model. We also show a Liouville theorem for the stationary solutions.
In this note, we study the phase transitions arising in a modified Smoluchowski equation on the sphere with dipolar potential. This equation models the competition between alignment and diffusion, and the modification consists in taking the strength of alignment and the intensity of the diffusion as functions of the order parameter. We characterize the stable and unstable equilibrium states. For stable equilibria, we provide the exponential rate of convergence. We detail special cases, giving rise to second order and first order phase transitions, respectively. We study the hysteresis diagram, and provide numerical illustrations of this phenomena.
Random walks have been successfully used to measure user or object similarities in collaborative filtering (CF) recommender systems, which is of high accuracy but low diversity. A key challenge of CF system is that the reliably accurate results are obtained with the help of peers' recommendation, but the most useful individual recommendations are hard to be found among diverse niche objects. In this paper we investigate the direction effect of the random walk on user similarity measurements and find that the user similarity, calculated by directed random walks, is reverse to the initial node's degree. Since the ratio of small-degree users to large-degree users is very large in real data sets, the large-degree users' selections are recommended extensively by traditional CF algorithms. By tuning the user similarity direction from neighbors to the target user, we introduce a new algorithm specifically to address the challenge of diversity of CF and show how it can be used to solve the accuracy-diversity dilemma. Without relying on any context-specific information, we are able to obtain accurate and diverse recommendations, which outperforms the state-of-the-art CF methods. This work suggests that the random walk direction is an important factor to improve the personalized recommendation performance.
Based on the user-item bipartite network, collaborative filtering (CF) recommender systems predict users' interests according to their history collections, which is a promising way to solve the information exploration problem. However, CF algorithm encounters cold start and sparsity problems. The trust-based CF algorithm is implemented by collecting the users' trust statements, which is time-consuming and must use users' private friendship information. In this paper, we present a novel measurement to calculate users' implicit trust-based correlation by taking into account their average ratings, rating ranges, and the number of common rated items. By applying the similar idea to the items, a item-based CF algorithm is constructed. The simulation results on three benchmark data sets show that the performances of both user-based and item-based algorithms could be enhanced greatly. Finally, a hybrid algorithm is constructed by integrating the user-based and item-based algorithms, the simulation results indicate that hybrid algorithm outperforms the state-of-the-art methods. Specifically, it can not only provide more accurate recommendations, but also alleviate the cold start problem.
Heat conduction process has recently found its application in personalized recommendation [T. Zhou \emph{et al.}, PNAS 107, 4511 (2010)], which is of high diversity but low accuracy. By decreasing the temperatures of small-degree objects, we present an improved algorithm, called biased heat conduction (BHC), which could simultaneously enhance the accuracy and diversity. Extensive experimental analyses demonstrate that the accuracy on MovieLens, Netflix and Delicious datasets could be improved by 43.5%, 55.4% and 19.2% compared with the standard heat conduction algorithm, and the diversity is also increased or approximately unchanged. Further statistical analyses suggest that the present algorithm could simultaneously identify users' mainstream and special tastes, resulting in better performance than the standard heat conduction algorithm. This work provides a creditable way for highly efficient information filtering.
We investigate systems of self-propelled particles with alignment interaction. Compared to previous work, the force acting on the particles is not normalized and this modification gives rise to phase transitions from disordered states at low density to aligned states at high densities. This model is the space inhomogeneous extension of a previous work by Frouvelle and Liu in which the existence and stability of the equilibrium states were investigated. When the density is lower than a threshold value, the dynamics is described by a non-linear diffusion equation. By contrast, when the density is larger than this threshold value, the dynamics is described by a hydrodynamic model for self-alignment interactions previously derived in Degond and Motsch. However, the modified normalization of the force gives rise to different convection speeds and the resulting model may lose its hyperbolicity in some regions of the state space.
This paper deals with the derivation and analysis of the the Hall Magneto-Hydrodynamic equations. We first provide a derivation of this system from a two-fluids Euler-Maxwell system for electrons and ions, through a set of scaling limits. We also propose a kinetic formulation for the Hall-MHD equations which contains as fluid closure different variants of the Hall-MHD model. Then, we prove the existence of global weak solutions for the incompressible viscous resistive Hall-MHD model. We use the particular structure of the Hall term which has zero contribution to the energy identity. Finally, we discuss particular solutions in the form of axisymmetric purely swirling magnetic fields and propose some regularization of the Hall equation.
This paper is concerned with the derivation and analysis of hydrodynamic models for systems of self-propelled particles subject to alignment interaction and attraction-repulsion. The starting point is the kinetic model considered in earlier work of Degond & Motsch with the addition of an attraction-repulsion interaction potential. Introducing different scalings than in Degond & Motsch, the non-local effects of the alignment and attraction-repulsion interactions can be kept in the hydrodynamic limit and result in extra pressure, viscosity terms and capillary force. The systems are shown to be symmetrizable hyperbolic systems with viscosity terms. A local-in-time existence result is proved in the 2D case for the viscous model and in the 3D case for the inviscid model. The proof relies on the energy method.
We consider a kinetic model of self-propelled particles with alignment interaction and with precession about the alignment direction. We derive a hydrodynamic system for the local density and velocity orientation of the particles. The system consists of the conservative equation for the local density and a non-conservative equation for the orientation. First, we assume that the alignment interaction is purely local and derive a first order system. However, we show that this system may lose its hyperbolicity. Under the assumption of weakly non-local interaction, we derive diffusive corrections to the first order system which lead to the combination of a heat flow of the harmonic map and Landau-Lifschitz-Gilbert dynamics. In the particular case of zero self-propelling speed, the resulting model reduces to the phenomenological Landau-Lifschitz-Gilbert equations. Therefore the present theory provides a kinetic formulation of classical micromagnetization models and spin dynamics.
In this paper we study the Euler-Poincaré equations in $\Bbb R^N$. We prove local existence of weak solutions in $W^{2,p}(\Bbb R^N),$ $p>N$, and local existence of unique classical solutions in $H^k (\Bbb R^N)$, $k>N/2+3$, as well as a blow-up criterion. For the zero dispersion equation($α=0$) we prove a finite time blow-up of the classical solution. We also prove that as the dispersion parameter vanishes, the weak solution converges to a solution of the zero dispersion equation with sharp rate as $α\to0$, provided that the limiting solution belongs to $C([0, T);H^k(\Bbb R^N))$ with $k>N/2 +3$. For the {\em stationary weak solutions} of the Euler-Poincaré equations we prove a Liouville type theorem. Namely, for $α>0$ any weak solution $\mathbf{u}\in H^1(\Bbb R^N)$ is $\mathbf{u}=0$; for $α=0$ any weak solution $\mathbf{u}\in L^2(\Bbb R^N)$ is $\mathbf{u}=0$.
Motivated by a phenomenon of phase transition in a model of alignment of self-propelled particles, we obtain a kinetic mean-field equation which is nothing else than the Doi equation (also called Smoluchowski equation) with dipolar potential. In a self-contained article, using only basic tools, we analyze the dynamics of this equation in any dimension. We first prove global well-posedness of this equation, starting with an initial condition in any Sobolev space. We then compute all possible steady-states. There is a threshold for the noise parameter: over this threshold, the only equilibrium is the uniform distribution, and under this threshold, there is also a family of non-isotropic equilibria. We give a rigorous prove of convergence of the solution to a steady-state as time goes to infinity. In particular we show that in the supercritical case, the only initial conditions leading to the uniform distribution in large time are those with vanishing momentum. For any positive value of the noise parameter, and any initial condition, we give rates of convergence towards equilibrium, exponentially for both supercritical and subcritical cases and algebraically for the critical case.
Recommender systems use data on past user preferences to predict possible future likes and interests. A key challenge is that while the most useful individual recommendations are to be found among diverse niche objects, the most reliably accurate results are obtained by methods that recommend objects based on user or object similarity. In this paper we introduce a new algorithm specifically to address the challenge of diversity and show how it can be used to resolve this apparent dilemma when combined in an elegant hybrid with an accuracy-focused algorithm. By tuning the hybrid appropriately we are able to obtain, without relying on any semantic or context-specific information, simultaneous gains in both accuracy and diversity of recommendations.
The dynamic environment in the real world calls for the adaptive techniques for information filtering, namely to provide real-time responses to the changes of system data. Where many incremental algorithms are designed for this purpose, they are usually challenged by the worse and worse performance resulted from the cumulative errors over time. In this Letter, we propose two incremental diffusion-based algorithms for the personalized recommendations, which integrate some pieces of local and fast updatings to achieve the approximate results. In addition to the fast responses, the errors of the proposed algorithms do not cumulate over time, that is to say, the global recomputing is unnecessary. This remarkable advantage is demonstrated by several metrics on algorithmic accuracy for two movie recommender systems and a social bookmarking system.
In this Letter, we introduce a modified collaborative filtering (MCF) algorithm, which has remarkably higher accuracy than the standard collaborative filtering. In the MCF, instead of the standard Pearson coefficient, the user-user similarities are obtained by a diffusion process. Furthermore, by considering the second order similarities, we design an effective algorithm that depresses the influence of mainstream preferences. The corresponding algorithmic accuracy, measured by the ranking score, is further improved by 24.9% in the optimal case. In addition, two significant criteria of algorithmic performance, diversity and popularity, are also taken into account. Numerical results show that the algorithm based on second order similarity can outperform the MCF simultaneously in all three criteria.
In this paper, we propose a spreading activation approach for collaborative filtering (SA-CF). By using the opinion spreading process, the similarity between any users can be obtained. The algorithm has remarkably higher accuracy than the standard collaborative filtering (CF) using Pearson correlation. Furthermore, we introduce a free parameter $β$ to regulate the contributions of objects to user-user correlations. The numerical results indicate that decreasing the influence of popular objects can further improve the algorithmic accuracy and personality. We argue that a better algorithm should simultaneously require less computation and generate higher accuracy. Accordingly, we further propose an algorithm involving only the top-$N$ similar neighbors for each target user, which has both less computational complexity and higher algorithmic accuracy.
In this paper, based on a weighted projection of the user-object bipartite network, we study the effects of user tastes on the mass-diffusion-based personalized recommendation algorithm, where a user's tastes or interests are defined by the average degree of the objects he has collected. We argue that the initial recommendation power located on the objects should be determined by both of their degree and the users' tastes. By introducing a tunable parameter, the user taste effects on the configuration of initial recommendation power distribution are investigated. The numerical results indicate that the presented algorithm could improve the accuracy, measured by the average ranking score, more importantly, we find that when the data is sparse, the algorithm should give more recommendation power to the objects whose degrees are close to the users' tastes, while when the data becomes dense, it should assign more power on the objects whose degrees are significantly different from user's tastes.
We present a mathematical analysis of the asymptotic preserving scheme proposed in [M. Lemou and L. Mieussens, SIAM J. Sci. Comput., 31, pp. 334-368, 2008] for linear transport equations in kinetic and diffusive regimes. We prove that the scheme is uniformly stable and accurate with respect to the mean free path of the particles. This property is satisfied under an explicitly given CFL condition. This condition tends to a parabolic CFL condition for small mean free paths, and is close to a convection CFL condition for large mean free paths. Ou r analysis is based on very simple energy estimates.
In this paper, by introducing a new user similarity index base on the diffusion process, we propose a modified collaborative filtering (MCF) algorithm, which has remarkably higher accuracy than the standard collaborative filtering. In the proposed algorithm, the degree correlation between users and objects is taken into account and embedded into the similarity index by a tunable parameter. The numerical simulation on a benchmark data set shows that the algorithmic accuracy of the MCF, measured by the average ranking score, is further improved by 18.19% in the optimal case. In addition, two significant criteria of algorithmic performance, diversity and popularity, are also taken into account. Numerical results show that the presented algorithm can provide more diverse and less popular recommendations, for example, when the recommendation list contains 10 objects, the diversity, measured by the hamming distance, is improved by 21.90%.