arXiv subjects

Zhou Fang

Publications and source records attributed to Zhou Fang.

At least 55 records · Page 3Linked to original sources

arXiv2022

Type-enriched Hierarchical Contrastive Strategy for Fine-Grained Entity Typing

Fine-grained entity typing (FET) aims to deduce specific semantic types of the entity mentions in text. Modern methods for FET mainly focus on learning what a certain type looks like. And few works directly model the type differences, that is, let models know the extent that one type is different from others. To alleviate this problem, we propose a type-enriched hierarchical contrastive strategy for FET. Our method can directly model the differences between hierarchical types and improve the ability to distinguish multi-grained similar types. On the one hand, we embed type into entity contexts to make type information directly perceptible. On the other hand, we design a constrained contrastive strategy on the hierarchical structure to directly model the type differences, which can simultaneously perceive the distinguishability between types at different granularity. Experimental results on three benchmarks, BBN, OntoNotes, and FIGER show that our method achieves significant performance on FET by effectively modeling type differences.

cs.CL↗

arXiv2022

Stochastic filtering for multiscale stochastic reaction networks based on hybrid approximations

In the past few decades, the development of fluorescent technologies and microscopic techniques has greatly improved scientists' ability to observe real-time single-cell activities. In this paper, we consider the filtering problem associate with these advanced technologies, i.e., how to estimate latent dynamic states of an intracellular multiscale stochastic reaction network from time-course measurements of fluorescent reporters. A good solution to this problem can further improve scientists' ability to extract information about intracellular systems from time-course experiments. A straightforward approach to this filtering problem is to use a particle filter where particles are generated by simulation of the full model and weighted according to observations. However, the exact simulation of the full dynamic model usually takes an impractical amount of computational time and prevents this type of particle filters from being used for real-time applications, such as transcription regulation networks. Inspired by the recent development of hybrid approximations to multiscale chemical reaction networks, we approach the filtering problem in an alternative way. We first prove that accurate solutions to the filtering problem can be constructed by solving the filtering problem for a reduced model that represents the dynamics as a hybrid process. The model reduction is based on exploiting the time-scale separations in the original network and, therefore, can greatly reduce the computational effort required to simulate the dynamics. As a result, we are able to develop efficient particle filters to solve the filtering problem for the original model by applying particle filters to the reduced model. We illustrate the accuracy and the computational efficiency of our approach using several numerical examples.

q-bio.QM↗

arXiv2022

Efficacy of regularized multi-task learning based on SVM models

This paper investigates the efficacy of a regularized multi-task learning (MTL) framework based on SVM (M-SVM) to answer whether MTL always provides reliable results and how MTL outperforms independent learning. We first find that M-SVM is Bayes risk consistent in the limit of large sample size. This implies that despite the task dissimilarities, M-SVM always produces a reliable decision rule for each task in terms of misclassification error when the data size is large enough. Furthermore, we find that the task-interaction vanishes as the data size goes to infinity, and the convergence rates of M-SVM and its single-task counterpart have the same upper bound. The former suggests that M-SVM cannot improve the limit classifier's performance; based on the latter, we conjecture that the optimal convergence rate is not improved when the task number is fixed. As a novel insight of MTL, our theoretical and experimental results achieved an excellent agreement that the benefit of the MTL methods lies in the improvement of the pre-convergence-rate factor (PCR, to be denoted in Section III) rather than the convergence rate. Moreover, this improvement of PCR factors is more significant when the data size is small.

stat.ML↗

arXiv2021

Homotopy Based Reinforcement Learning with Maximum Entropy for Autonomous Air Combat

The Intelligent decision of the unmanned combat aerial vehicle (UCAV) has long been a challenging problem. The conventional search method can hardly satisfy the real-time demand during high dynamics air combat scenarios. The reinforcement learning (RL) method can significantly shorten the decision time via using neural networks. However, the sparse reward problem limits its convergence speed and the artificial prior experience reward can easily deviate its optimal convergent direction of the original task, which raises great difficulties for the RL air combat application. In this paper, we propose a homotopy-based soft actor-critic method (HSAC) which focuses on addressing these problems via following the homotopy path between the original task with sparse reward and the auxiliary task with artificial prior experience reward. The convergence and the feasibility of this method are also proved in this paper. To confirm our method feasibly, we construct a detailed 3D air combat simulation environment for the RL-based methods training firstly, and we implement our method in both the attack horizontal flight UCAV task and the self-play confrontation task. Experimental results show that our method performs better than the methods only utilizing the sparse reward or the artificial prior experience reward. The agent trained by our method can reach more than 98.3% win rate in the attack horizontal flight UCAV task and average 67.4% win rate when confronted with the agents trained by the other two methods.

cs.LG↗

arXiv2021

Convergence of regularized particle filters for stochastic reaction networks

Filtering for stochastic reaction networks (SRNs) is an important problem in systems/synthetic biology aiming to estimate the state of unobserved chemical species. A good solution to it can provide scientists valuable information about the hidden dynamic state and enable optimal feedback control. Usually, the model parameters need to be inferred simultaneously with state variables, and a conventional particle filter can fail to solve this problem accurately due to sample degeneracy. In this case, the regularized particle filter (RPF) is preferred to the conventional ones, as the RPF can mitigate sample degeneracy by perturbing particles with artificial noise. However, the artificial noise introduces an additional bias to the estimate, and, thus, it is questionable whether the RPF can provide reliable results for SRNs. In this paper, we aim to identify conditions under which the RPF converges to the exact filter in the filtering problem determined by a bimolecular network. First, we establish computationally efficient RPFs for SRNs on different scales using different dynamical models, including the continuous-time Markov process, tau-leaping model, and piecewise deterministic process. Then, by parameter sensitivity analyses, we show that the established RPFs converge to the exact filters if all reactions leading to an increase of the molecular population have linearly growing propensities and some other mild conditions are satisfied simultaneously. This ensures the performance of the RPF for a large class of SRNs, and several numerical examples are presented to illustrate our results.

q-bio.QM↗

arXiv2021

Promoting Mental Well-Being for Audiences in a Live-Streaming Game by Highlight-Based Bullet Comments

This paper proposes a method for generating bullet comments for live-streaming games based on highlights (i.e., the exciting parts of video clips) extracted from the game content and evaluate the effect of mental health promotion. Game live streaming is becoming a popular theme for academic research. Compared to traditional online video sharing platforms, such as Youtube and Vimeo, video live streaming platform has the benefits of communicating with other viewers in real-time. In sports broadcasting, the commentator plays an essential role as mood maker by making matches more exciting. The enjoyment emerged while watching game live streaming also benefits the audience's mental health. However, many e-sports live streaming channels do not have a commentator for entertaining viewers. Therefore, this paper presents a design of an AI commentator that can be embedded in live streaming games. To generate bullet comments for real-time game live streaming, the system employs highlight evaluation to detect the highlights, and generate the bullet comments. An experiment is conducted and the effectiveness of generated bullet comments in a live-streaming fighting game channel is evaluated.

cs.MM↗

arXiv2021

Fighting Game Commentator with Pitch and Loudness Adjustment Utilizing Highlight Cues

This paper presents a commentator for providing real-time game commentary in a fighting game. The commentary takes into account highlight cues, obtained by analyzing scenes during gameplay, as input to adjust the pitch and loudness of commentary to be spoken by using a Text-to-Speech (TTS) technology. We investigate different designs for pitch and loudness adjustment. The proposed AI consists of two parts: a dynamic adjuster for controlling pitch and loudness of the TTS and a real-time game commentary generator. We conduct a pilot study on a fighting game, and our result shows that by adjusting the loudness significantly according to the level of game highlight, the entertainment of the gameplay can be enhanced.

cs.MM↗

arXiv2020

DeepStreet: A deep learning powered urban street network generation module

In countries experiencing unprecedented waves of urbanization, there is a need for rapid and high quality urban street design. Our study presents a novel deep learning powered approach, DeepStreet (DS), for automatic street network generation that can be applied to the urban street design with local characteristics. DS is driven by a Convolutional Neural Network (CNN) that enables the interpolation of streets based on the areas of immediate vicinity. Specifically, the CNN is firstly trained to detect, recognize and capture the local features as well as the patterns of the existing street network sourced from the OpenStreetMap. With the trained CNN, DS is able to predict street networks' future expansion patterns within the predefined region conditioned on its surrounding street networks. To test the performance of DS, we apply it to an area in and around the Eixample area in the City of Barcelona, a well known example in the fields of urban and transport planning with iconic grid like street networks in the centre and irregular road alignments farther afield. The results show that DS can (1) detect and self cluster different types of complex street patterns in Barcelona; (2) predict both gridiron and irregular street and road networks. DS proves to have a great potential as a novel tool for designers to efficiently design the urban street network that well maintains the consistency across the existing and newly generated urban street network. Furthermore, the generated networks can serve as a benchmark to guide the local plan-making especially in rapidly developing cities.

cs.CV↗

arXiv2020

Incorporating planning intelligence into deep learning: A planning support tool for street network design

Deep learning applications in shaping ad hoc planning proposals are limited by the difficulty in integrating professional knowledge about cities with artificial intelligence. We propose a novel, complementary use of deep neural networks and planning guidance to automate street network generation that can be context-aware, example-based and user-guided. The model tests suggest that the incorporation of planning knowledge (e.g., road junctions and neighborhood types) in the model training leads to a more realistic prediction of street configurations. Furthermore, the new tool provides both professional and lay users an opportunity to systematically and intuitively explore benchmark proposals for comparisons and further evaluations.

cs.CV↗

arXiv2020

A graphic formulation of non-isothermal chemical reaction systems and the analysis of detailed balanced networks

In this paper, we provide a graphic formulation of non-isothermal reaction systems and show that a non-isothermal detailed balanced network system converges (locally) asymptotically to the unique equilibrium within the invariant manifold determined by the initial condition. To model thermal effects, the proposed modeling approach extends the classical chemical reaction network by adding two parameters to each direct (reaction) edge, depicting, respectively, the instantaneous internal energy change after the firing of the reaction and the variation of the reaction rate with respect to the temperature. For systems possessing thermodynamic equilibria, our modeling approach provides a compact formulation of the dynamics where reaction topology and thermodynamic information are presented simultaneously. Finally, using this formulation and the Legendre transformation, we show that non-isothermal detailed balanced network systems admit some fundamental properties: dissipativeness, the detailed balancing of each equilibrium, the existence and uniqueness of the equilibrium, and the asymptotic stability of each equilibrium. In general, the analysis and results of this work provide insights into the research of non-isothermal chemical reaction systems.

math.DS↗

arXiv2020

Stochastic filters based on hybrid approximations of multiscale stochastic reaction networks

We consider the problem of estimating the dynamic latent states of an intracellular multiscale stochastic reaction network from time-course measurements of fluorescent reporters. We first prove that accurate solutions to the filtering problem can be constructed by solving the filtering problem for a reduced model that represents the dynamics as a hybrid process. The model reduction is based on exploiting the time-scale separations in the original network, and it can greatly reduce the computational effort required to simulate the dynamics. This enables us to develop efficient particle filters to solve the filtering problem for the original model by applying particle filters to the reduced model. We illustrate the accuracy and the computational efficiency of our approach using a numerical example.

stat.ME↗

arXiv2020

GP-SLAM+: real-time 3D lidar SLAM based on improved regionalized Gaussian process map reconstruction

This paper presents a 3D lidar SLAM system based on improved regionalized Gaussian process (GP) map reconstruction to provide both low-drift state estimation and mapping in real-time for robotics applications. We utilize spatial GP regression to model the environment. This tool enables us to recover surfaces including those in sparsely scanned areas and obtain uniform samples with uncertainty. Those properties facilitate robust data association and map updating in our scan-to-map registration scheme, especially when working with sparse range data. Compared with previous GP-SLAM, this work overcomes the prohibitive computational complexity of GP and redesigns the registration strategy to meet the accuracy requirements in 3D scenarios. For large-scale tasks, a two-thread framework is employed to suppress the drift further. Aerial and ground-based experiments demonstrate that our method allows robust odometry and precise mapping in real-time. It also outperforms the state-of-the-art lidar SLAM systems in our tests with light-weight sensors.

cs.RO↗

arXiv2020

Time-domain Boundedness of Noise-to-State Exponentially Stable Systems

In this paper we prove the time-domain boundedness for noise-to-state exponentially stable systems, and further make an estimation of its lower bound function, which allows to answer the question that how long the solution of a stochastic noise-to-state exponentially stable system stays in the domain of attraction and what happens with it if it escapes from this region for a while. The results will complement the probability-domain boundedness of noise-to-state exponentially stable systems, and provide a new insight into noise-to-state exponential stability.

math.DS↗

arXiv2019

Towards An Angry-Birds-like Game System for Promoting Mental Well-being of Players Using Art-Therapy-embedded PCG

This paper presents an integration of a game system and the art therapy concept for promoting the mental well-being of video game players. In the proposed game system, the player plays an Angry-Birds-like game in which levels in the game are generated based on images they draw. Upon finishing a game level, the player also receives positive feedback (praising words) toward their drawing and the generated level from an Art Therapy AI. The proposed system is composed of three major parts: (1) a drawing recognizer that identifies what object is drawn by the player (Sketcher), (2) a level generator that converts the drawing image into a pixel image, then a set of blocks representing a game level (PCG AI), and (3) the Art Therapy AI that encourages the player and improves their emotion. This paper describes an overview of the system and explains how its major components function.

cs.HC↗

arXiv2019

Lyapunov Function Partial Differential Equations for Chemical Reaction Networks: Some Special Cases

In this paper, we develop a method to generate the Lyapunov function for stability analysis for chemical reaction networks. Based on the Chemical Master Equation, we derive the Lyapunov Function partial differential equations (PDEs), whose solution approximates the scaling non-equilibrium potential and serves as the candidate Lyapunov function for the given network. We further prove that for any chemical reaction network the solution (if exists) of the PDEs is dissipative. Moreover, the proposed method of Lyapunov Function PDEs is qualified for analyzing the asymptotic stability of complex balanced networks, all networks with $1$-dimensional stoichiometric subspace and some special networks with more than $2$-dimensional stoichiometric subspace if some moderate conditions are added. Several examples are presented to illustrate the efficiency of the method.

math.DS↗

arXiv2019

Revisiting Persistence of Chemical Reaction Networks through Lyapunov Function Partial Differential Equations

In this paper, we propose a novel technique, referred to as the Lyapunov Function PDEs technique, to diagnose persistence of chemical reaction networks with mass-action kinetics. The technique allows that every network is attached to a Lyapuonv function PDE and a boundary condition whose solutions are expected to be Lyapunov functions for the network. By means of solution of the PDEs, either in the forms of itself or its time derivative, some checkable criteria are proposed for persistence of network systems. These criteria show high validity in proving that neither non-semilocking boundary points nor semilocking boundary non-equilibrium points (with additional conditions included) in mass-action systems are $ω$-$limit$ points. Further, we prove that a class of networks, called $\mathcal{W}_{I}$-endotactic networks that are regardless of values of the system parameters, a set larger than endotactic networks but smaller than $w_{I}$-endotactic networks, also give rise to persistent systems if the networks are $1$-dimensional. Although part of our results are covered by existing ones, part of them are still new. The proposed Lyapunov Function PDEs technique allows us to check persistence of mass action systems in an alternative corner, and is exhibiting large potential.

math.DS↗

arXiv2018

Complex Balancing Reconstructed to the Asymptotic Stability of Mass-action Chemical Reaction Networks with Conservation Laws

Motivated by the fact that the pseudo-Helmholtz function is a valid Lyapunov function for characterizing asymptotic stability of complex balanced mass action systems (MASs), this paper develops the generalized pseudo-Helmholtz function for stability analysis for more general MASs assisted with conservation laws. The key technique is to transform the original network into two different MASs, defined by reconstruction and reverse reconstruction, with an important aspect that the dynamics of the original network for free species is equivalent to that of the reverse reconstruction. Stability analysis of the original network is then conducted based on an analysis of how stability properties are retained from the original network to the reverse reconstruction. We prove that the reverse reconstruction possesses only an equilibrium in each positive stoichiometric compatibility class if the corresponding reconstruction is complex balanced. Under this complex balanced reconstruction strategy, the asymptotic stability of the reverse reconstruction, which also applies to the original network, is thus reached by taking the generalized pseudo-Helmholtz function as the Lyapunov function. To facilitate applications, we further provide a systematic method for computing complex balanced reconstructions assisted with conservation laws. Some representative examples are presented to exhibit the validity of the complex balanced reconstruction strategy.

math.DS↗

arXiv2017

Stabilization of (state, input)-disturbed CSTRs through the port-Hamiltonian systems approach

It is a universal phenomenon that the state and input of the continuous stirred tank reactor (CSTR) systems are both disturbed. This paper proposes a (state, input)-disturbed port-Hamiltonian framework that can be used to model and further designs a stochastic passivity based controller to asymptotically stabilize in probability the (state, input)-disturbed CSTR (sidCSTR) systems. The opposite entropy function and the availability function are selected as the Hamiltonian for the model and control purposes, respectively. Furthermore, the proposed (state, input)-disturbed port-Hamiltonian model can simultaneously characterize the first law and the second law of thermodynamics when the opposite entropy function acts as the Hamiltonian. A simple CSTR example illustrates how the port-Hamiltonian method is utilized for modeling and controlling sidCSTR systems.

math.OC↗