arXiv subjects

Chuanhou Gao

Publications and source records attributed to Chuanhou Gao.

At least 37 records · Page 2Linked to original sources

arXiv2023

Domain Knowledge integrated for Blast Furnace Classifier Design

Blast furnace modeling and control is one of the important problems in the industrial field, and the black-box model is an effective mean to describe the complex blast furnace system. In practice, there are often different learning targets, such as safety and energy saving in industrial applications, depending on the application. For this reason, this paper proposes a framework to design a domain knowledge integrated classification model that yields a classifier for industrial application. Our knowledge incorporated learning scheme allows the users to create a classifier that identifies "important samples" (whose misclassifications can lead to severe consequences) more correctly, while keeping the proper precision of classifying the remaining samples. The effectiveness of the proposed method has been verified by two real blast furnace datasets, which guides the operators to utilize their prior experience for controlling the blast furnace systems better.

cs.LG↗

arXiv2023

Capturing persistence of delayed complex balanced chemical reaction systems via decomposition of semilocking sets

With the increasing complexity of time-delayed systems, the diversification of boundary types of chemical reaction systems poses a challenge for persistence analysis. This paper focuses on delayed complex balanced mass action systems (DeCBMAS) and derives that some boundaries of a DeCBMAS can not contain an $ω$-limit point of some trajectory with positive initial point by using the method of semilocking set decomposition and the property of the facet, further expanding the range of persistence of delayed complex balanced systems. These findings demonstrate the effectiveness of semilocking set decomposition to address the complex boundaries and offer insights into the persistence analysis of delayed chemical reaction network systems.

math.DS↗

arXiv2023

Chemical relaxation oscillator designed to control molecular computation

Embedding efficient calculation instructions into biochemical system has always been a research focus in synthetic biology. One of the key problems is how to sequence the chemical reaction modules that act as units of computation and make them alternate spontaneously. Our work takes the design of chemical clock signals as a solution and presents a $4$-dimensional chemical oscillator model based on relaxation oscillation to generate a pair of symmetric clock signals for two-module regulation. We give detailed dynamical analysis of the model and discuss how to control the period and occurrence order of clock signals. We also demonstrate the loop control of molecular computations and provide termination strategy for them. We can expect that our design for module regulation and loop termination will help advance the embedding of more complicate calculations into biochemical environments.

math.DS↗

arXiv2022

Accurate control to run and stop chemical reactions via relaxation oscillators

Regulation of multiple reaction modules is quite common in molecular computation and deep learning networks construction through chemical reactions, as is always a headache for that sequential execution of modules goes against the intrinsically parallel nature of chemical reactions. Precisely switching multiple reaction modules both on and off acts as the core role in programming chemical reaction systems. Unlike setting up physical compartments or adding human intervention signals, we adopt the idea of chemical oscillators based on relaxation oscillation, and assign corresponding clock signal components into the modules that need to be regulated. This paper mainly demonstrates the design process of oscillators under the regulation task of three modules, and provides a suitable approach for automatic termination of the modules cycle. We provide the simulation results at the level of ordinary differential equation and ensure that equations can be translated into corresponding chemical reaction networks.

math.DS↗

arXiv2022

ELBD: Efficient score algorithm for feature selection on latent variables of VAE

In this paper, we develop the notion of evidence lower bound difference (ELBD), based on which an efficient score algorithm is presented to implement feature selection on latent variables of VAE and its variants. Further, we propose weak convergence approximation algorithms to optimize VAE related models through weighing the ``more important" latent variables selected and accordingly increasing evidence lower bound. We discuss two kinds of different Gaussian posteriors, mean-filed and full-covariance, for latent variables, and make corresponding theoretical analyses to support the effectiveness of algorithms. A great deal of comparative experiments are carried out between our algorithms and other 9 feature selection methods on 7 public datasets to address generative tasks. The results provide the experimental evidence of effectiveness of our algorithms. Finally, we extend ELBD to its generalized version, and apply the latter to tackling classification tasks of 5 new public datasets with satisfactory experimental results.

stat.ML↗

arXiv2022

Design of universal chemical relaxation oscillator to control molecular computation

Embedding efficient command operation into biochemical system has always been a research focus in synthetic biology. One of the key problems is how to sequence the chemical reactions that act as units of computation. The answer is to design chemical oscillator, a component that acts as a clock signal to turn corresponding reaction on or off. Some previous work mentioned the use of chemical oscillations. However, the models used either lack a systematic analysis of the mechanism and properties of oscillation, or are too complex to be tackled with in practice. Our work summarizes the universal process for designing chemical oscillators, including generating robust oscillatory species, constructing clock signals from these species, and setting up termination component to eventually end the loop of whole reaction modules. We analyze the dynamic properties of the proposed oscillator model in the context of ordinary differential equations, and discuss how to determine parameters for the effect we want in detail. Our model corresponds to abstract chemical reactions based on mass-action kinetics which are expected to be implemented into chemistry with the help of DNA strand displacement cascades. Our consideration of ordering chemical reaction modules helps advance the embedding of more complex calculations into biochemical environments.

math.DS↗

arXiv2022

Use statistical analysis to approximate integrated order batching problem

Order picking and order packing entail retrieving items from storage and packaging them according to customer requests. These activities have always been the main concerns of the companies in reducing warehouse management costs. This paper proposes and investigates the Order Batching and Order Packing Problem, which considers these activities jointly. The authors propose a novel statistic-based framework, namely, the Max Correlation Reformulation problem, to find an approximation mixed-integer programming model. An approximation model is found within this framework in two phases. A lower dimension model is firstly proposed. Efforts are then made to increase its correlation coefficient with the original formulation. Finally, a powerful pairs swapping heuristics is combined with the approximation model. Numerical experiments show that this newly found approach outperforms the mainstream methods from the literature. It is demonstrated that this proposed method could significantly reduce the cost of picking and packing operations in a warehouse.

math.OC↗

arXiv2022

Transfer Learning in Information Criteria-based Feature Selection

This paper investigates the effectiveness of transfer learning based on Mallows' Cp. We propose a procedure that combines transfer learning with Mallows' Cp (TLCp) and prove that it outperforms the conventional Mallows' Cp criterion in terms of accuracy and stability. Our theoretical results indicate that, for any sample size in the target domain, the proposed TLCp estimator performs better than the Cp estimator by the mean squared error (MSE) metric in the case of orthogonal predictors, provided that i) the dissimilarity between the tasks from source domain and target domain is small, and ii) the procedure parameters (complexity penalties) are tuned according to certain explicit rules. Moreover, we show that our transfer learning framework can be extended to other feature selection criteria, such as the Bayesian information criterion. By analyzing the solution of the orthogonalized Cp, we identify an estimator that asymptotically approximates the solution of the Cp criterion in the case of non-orthogonal predictors. Similar results are obtained for the non-orthogonal TLCp. Finally, simulation studies and applications with real data demonstrate the usefulness of the TLCp scheme.

stat.ML↗

arXiv2022

Gaussian mixture modeling of nodes in Bayesian network according to maximal parental cliques

This paper uses Gaussian mixture model instead of linear Gaussian model to fit the distribution of every node in Bayesian network. We will explain why and how we use Gaussian mixture models in Bayesian network. Meanwhile we propose a new method, called double iteration algorithm, to optimize the mixture model, the double iteration algorithm combines the expectation maximization algorithm and gradient descent algorithm, and it performs perfectly on the Bayesian network with mixture models. In experiments we test the Gaussian mixture model and the optimization algorithm on different graphs which is generated by different structure learning algorithm on real data sets, and give the details of every experiment.

stat.ML↗

arXiv2022

Towards Programming Adaptive Linear Neural Networks Through Chemical Reaction Networks

This paper is concerned with programming adaptive linear neural networks (ALNNs) using chemical reaction networks (CRNs) equipped with mass-action kinetics. Through individually programming the forward propagation and the backpropagation of ALNNs, and also utilizing the permeation walls technique, we construct a powerful CRN possessing the function of ALNNs, especially having the function of automatic computation. We also provide theoretical analysis and a case study to support our construction. The results will have potential implications for the developments of synthetic biology, molecular computer and artificial intelligence.

math.DS↗

arXiv2022

On Stability of Two Kinds of Delayed Chemical Reaction Networks

For the networks that are linear conjugate to complex balanced systems, the delayed version may include two classes of networks: one class is still linear conjugate to the delayed complex balanced network, the other is not. In this paper, we prove the existence of the first class of networks, and emphasize the local asymptotic stability relative to a certain defined invariant set. For the second class of systems, we define a special subclass and derive the local asymptotic stability for the subclass. Two examples are provided to illustrate our results.

math.DS↗

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

Online Interaction Detection for Click-Through Rate Prediction

Click-Through Rate prediction aims to predict the ratio of clicks to impressions of a specific link. This is a challenging task since (1) there are usually categorical features, and the inputs will be extremely high-dimensional if one-hot encoding is applied, (2) not only the original features but also their interactions are important, (3) an effective prediction may rely on different features and interactions in different time periods. To overcome these difficulties, we propose a new interaction detection method, named Online Random Intersection Chains. The method, which is based on the idea of frequent itemset mining, detects informative interactions by observing the intersections of randomly chosen samples. The discovered interactions enjoy high interpretability as they can be comprehended as logical expressions. ORIC can be updated every time new data is collected, without being retrained on historical data. What's more, the importance of the historical and latest data can be controlled by a tuning parameter. A framework is designed to deal with the streaming interactions, so almost all existing models for CTR prediction can be applied after interaction detection. Empirical results demonstrate the efficiency and effectiveness of ORIC on three benchmark datasets.

cs.LG↗

arXiv2021

Random Intersection Chains

Interactions between several features sometimes play an important role in prediction tasks. But taking all the interactions into consideration will lead to an extremely heavy computational burden. For categorical features, the situation is more complicated since the input will be extremely high-dimensional and sparse if one-hot encoding is applied. Inspired by association rule mining, we propose a method that selects interactions of categorical features, called Random Intersection Chains. It uses random intersections to detect frequent patterns, then selects the most meaningful ones among them. At first a number of chains are generated, in which each node is the intersection of the previous node and a random chosen observation. The frequency of patterns in the tail nodes is estimated by maximum likelihood estimation, then the patterns with largest estimated frequency are selected. After that, their confidence is calculated by Bayes' theorem. The most confident patterns are finally returned by Random Intersection Chains. We show that if the number and length of chains are appropriately chosen, the patterns in the tail nodes are indeed the most frequent ones in the data set. We analyze the computation complexity of the proposed algorithm and prove the convergence of the estimators. The results of a series of experiments verify the efficiency and effectiveness of the algorithm.

stat.ML↗

arXiv2021

Discovering Categorical Main and Interaction Effects Based on Association Rule Mining

With the growing size of data sets, feature selection becomes increasingly important. Taking interactions of original features into consideration will lead to extremely high dimension, especially when the features are categorical and one-hot encoding is applied. This makes it more worthwhile mining useful features as well as their interactions. Association rule mining aims to extract interesting correlations between items, but it is difficult to use rules as a qualified classifier themselves. Drawing inspiration from association rule mining, we come up with a method that uses association rules to select features and their interactions, then modify the algorithm for several practical concerns. We analyze the computational complexity of the proposed algorithm to show its efficiency. And the results of a series of experiments verify the effectiveness of the algorithm.

cs.DB↗

arXiv2021

Chemical Reaction Network Decomposition Technique for Stability Analysis

This paper develops the concept of decomposition for chemical reaction networks, based on which a network decomposition technique is proposed to capture the stability of large-scale networks characterized by a high number of species, high dimension, high deficiency, and/or non-weakly reversible structure. We present some sufficient conditions to capture the stability of a network (may possess any dimension, any deficiency, and/or any topological structure) when it can be decomposed into a complex balanced subnetwork and a few 1-dimensional subnetworks (and/or a few two-species subnetworks), especially in the case when there are shared species in different subnetworks. The results cover encouraging applications on autocatalytic networks with some frequently-encountered biochemical reactions examples of interest, such as the autophosphorylation of PAK1 and Aurora B kinase, autocatalytic cycles originating from metabolism, etc.

math.DS↗

arXiv2020

Structured Sparsity Modeling for Improved Multivariate Statistical Analysis based Fault Isolation

In order to improve the fault diagnosis capability of multivariate statistical methods, this article introduces a fault isolation framework based on structured sparsity modeling. The developed method relies on the reconstruction based contribution analysis and the process structure information can be incorporated into the reconstruction objective function in the form of structured sparsity regularization terms. The structured sparsity terms allow selection of fault variables over structures like blocks or networks of process variables, hence more accurate fault isolation can be achieved. Four structured sparsity terms corresponding to different kinds of process information are considered, namely, partially known sparse support, block sparsity, clustered sparsity and tree-structured sparsity. The optimization problems involving the structured sparsity terms can be solved using the Alternating Direction Method of Multipliers (ADMM) algorithm, which is fast and efficient. Through a simulation example and an application study to a coal-fired power plant, it is verified that the proposed method can better isolate faulty variables by incorporating process structure information.

stat.AP↗

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↗