Alice Chen

• The University of Hong Kong

Deep learning based semi-supervised learning (SSL) algorithms have led to promising results in recent years. However, they tend to introduce multiple tunable hyper-parameters, making them less practical in real SSL scenarios where the labeled data is scarce for extensive hyper-parameter search. In this paper, we propose a novel meta-learning based...

Few-shot learning aims to make classification when few samples are available. In general, metric-based methods map images into a space by learning the embedding function. However, conventional metric-based methods rely on a single distance value, which does not pay attention to the shallow features. In this paper, we propose a multi-distance metric...

Wearable devices have dramatically developed over the past decade as their functions extended from the simple posture analysis to non-invasive condition monitoring for early warning and proactive healthcare, which are especially significant for the dangerous disease such as cardiac arrhythmia. However, it is difficult for the wearable devices to co...

Idiopathic scoliosis (IS) is a common lifetime disease, which exhibits an obvious deformity of spinal curvature to seriously affect heart and lung function. Accurate radiographic assessment of spinal curvature is vitally important for the clinical diagnosis and treatment planning of idiopathic scoliosis. Deep learning algorithms have been widely ad...

In this article, we focus on a biobjective hot strip mill (HSM) scheduling problem arising in the steel industry. Besides the conventional objective regarding penalty costs, we have also considered minimizing the total starting times of rolling operations in order to reduce the energy consumption for slab reheating. The problem is complicated by th...

Deep reinforcement learning (RL) agents are becoming increasingly proficient in a range of complex control tasks. However, the agent's behavior is usually difficult to interpret due to the introduction of black-box function, making it difficult to acquire the trust of users. Although there have been some interesting interpretation methods for visio...

Deep reinforcement learning (RL) agents are becoming increasingly proficient in a range of complex control tasks. However, the agent's behavior is usually difficult to interpret due to the introduction of black-box function, making it difficult to acquire the trust of users. Although there have been some interesting interpretation methods for visio...

Although considerable success has been achieved in urban air quality prediction (AQP) with machine learning techniques, accurate and long-term prediction is still challenging. One of the key issues for existing AQP approaches is that air quality monitoring stations are sparsely distributed, typically with around ten monitoring stations per city. As...

Wearable devices have dramatically developed over the past decade as their functions extended from the simple posture analysis to non-invasive condition monitoring for early warning and proactive healthcare, which are especially significant for the dangerous disease such as cardiac arrhythmia. However, it is difficult for the wearable devices to co...

Data augmentation is widely known as a simple yet surprisingly effective technique for regularizing deep networks. In this paper, we propose a novel semantic data augmentation algorithm to complement traditional schemes, such as flipping, translation and rotation. The proposed method is inspired by the intriguing property that deep networks are eff...

Deep reinforcement learning (RL) has recently led to many breakthroughs on a range of complex control tasks. However, the decision-making process is generally not transparent. The lack of interpretability hinders the applicability in safety-critical scenarios. While several methods have attempted to interpret vision-based RL, most come without deta...

Data augmentation is widely known as a simple yet surprisingly effective technique for regularizing deep networks. Conventional data augmentation schemes, e.g., flipping, translation or rotation, are low-level, data-independent and class-agnostic operations, leading to limited diversity for augmented samples. To this end, we propose a novel semanti...

Deep neural networks (DNNs) have been very successful for supervised learning. However, their high generalization performance often comes with the high cost of annotating data manually. Collecting low-quality labeled dataset is relatively cheap, e.g., using web search engines, while DNNs tend to overfit to corrupted labels easily. In this paper, we...

Scheduling of hot strip mill is an important decision problem in the steel manufacturing industry. Previous studies on the hot strip mill scheduling problem have mostly neglected the random factors in production. However, random variations in processing times are inevitable due to unpredictable delays and disturbances. In this paper, we adopt a rob...

In this paper, we propose a novel implicit semantic data augmentation (ISDA) approach to complement traditional augmentation techniques like flipping, translation or rotation. Our work is motivated by the intriguing property that deep networks are surprisingly good at linearizing features, such that certain directions in the deep feature space corr...

Maximum entropy deep reinforcement learning (RL) methods have been demonstrated on a range of challenging continuous tasks. However, existing methods either suffer from severe instability when training on large off-policy data or cannot scale to tasks with very high state and action dimensionality such as 3D humanoid locomotion. Besides, the optima...

Model-free deep reinforcement learning (RL) algorithms have been widely used for a range of complex control tasks. However, slow convergence and sample inefficiency remain challenging problems in RL, especially when handling continuous and high-dimensional state spaces. To tackle this problem, we propose a general acceleration method for model-free...

This paper investigates trajectory tracking problem for a class of underactuated autonomous underwater vehicles (AUVs) with unknown dynamics and constrained inputs. Different from existing policy gradient methods which employ single actor-critic but cannot realize satisfactory tracking control accuracy and stable learning, our proposed algorithm ca...

Maximum entropy deep reinforcement learning (RL) methods have been demonstrated on a range of challenging continuous tasks. However, existing methods either suffer from severe instability when training on large off-policy data or cannot scale to tasks with very high state and action dimensionality such as 3D humanoid locomotion. Besides, the optima...

Domain adaptation aims to deal with learning problems in which the labeled training data and unlabeled testing data are differently distributed. Maximum Mean Discrepancy (MMD), as a distribution distance measure, is minimized in various domain adaptation algorithms for eliminating domain divergence. We analyze empirical MMD from the point of view o...

Localizing target nodes is a fundamental problem for wireless sensor networks (WSNs). Without range measurements, the range-free techniques, which only exploit the connectivity information among nodes, have been widely studied in the past score years. How to achieve a good balance between the localization accuracy and the communication cost has not...

In this brief, we propose a novel nonparametric supervised linear dimension reduction (SLDR) algorithm that extracts the features by maximizing the pairwise separation probability. The separation probability, as a new class separability measure, describes the generalization accuracy when we use the obtained features to train a linear classifier. Ob...

This paper investigates trajectory tracking problem for a class of underactuated autonomous underwater vehicles (AUVs) with unknown dynamics and constrained inputs. Different from existing policy gradient methods which employ single actor critic but cannot realize satisfactory tracking control accuracy and stable learning, our proposed algorithm ca...

Domain adaptation manages to build an effective target classifier or regression model for unlabeled target data by utilizing the well-labeled source data but lying different distributions. Intuitively, to address domain shift problem, it is crucial to learn domain invariant features across domains, and most existing approaches have concentrated on...

Extreme learning machine (ELM) has been applied in a wide range of classification and regression problems due to its high accuracy and efficiency. However, ELM can only deal with cases where training and testing data are from identical distribution, while in real world situations, this assumption is often violated. As a result, ELM performs poorly...

Extreme learning machines (ELMs), as "generalized" single hidden layer feedforward networks, have been proved to be effective and efficient for classification and regression problems. Traditional ELMs assume that the training and testing data are drawn from the same distribution, which however is often violated in real-world applications. In this p...

This paper studies problems on locally stopping distributed consensus algorithms over networks where each node updates its state by interacting with its neighbors and decides by itself whether certain level of agreement has been achieved among nodes. Since an individual node is unable to access the states of those beyond its neighbors, this problem...

In this paper, we consider depth control problems of an autonomous underwater vehicle (AUV) for tracking the desired depth trajectories. Due to the unknown dynamical model of the AUV, the problems cannot be solved by most of model-based controllers. To this purpose, we formulate the depth control problems of the AUV as continuous-state, continuous-...

In this paper, we consider depth control problems of an autonomous underwater vehicle (AUV) for tracking the desired depth trajectories. Due to the unknown dynamical model of the AUV, the problems cannot be solved by most of model-based controllers. To this purpose, we formulate the depth control problems of the AUV as continuous-state, continuous-...

Accurate ore grade estimation is crucial to mineral resources evaluation and exploration. In this paper, we consider the borehole data collected from the Solwara 1 deposit, where the hydrothermal sulfide ore body is quite complicated with incomplete ore grade values. To solve this estimation problem, the relevance vector machine (RVM) and the expec...

Textile dyeing often constitutes a bottleneck procedure in the production of clothing because the dyeing process is time-consuming and heavily constrained. Meanwhile, dyeing processes inevitably produce emissions of water pollutants especially when the involved equipment undergoes cleaning operations. Scheduling could be utilized as a system-level...

This paper is concerned with a constrained optimization problem over a directed graph (digraph) of nodes, in which the cost function is a sum of local objectives, and each node only knows its local objective and constraints. To collaboratively solve the optimization, most of the existing works require the interaction graph to be balanced or " doubl...

This paper is concerned with a constrained optimization problem over a directed graph (digraph) of nodes, in which the cost function is a sum of local objectives, and each node only knows its local objective and constraints. To collaboratively solve the optimization, most of the existing works require the interaction graph to be balanced or “doubly...

Routing service considering uncertainty is at the core of intelligent transportation systems and has attracted increasing attention. Existing stochastic shortest path models require the exact probability distributions of travel times and usually assume that they are independent. However, the distributions are often unavailable or inaccurate due to...

We discuss stochastic multi-item capacitated lot-sizing problems with and without setup carryovers (also known as link lot size), S-MICLSP and S-MICLSP-L. The two models are motivated from a real-world steel enterprise. To overcome the nonlinearity of the models, a piecewise linear approximation method is proposed. We develop a new fix-and-optimize...

This paper studies problems on locally stopping distributed consensus algorithms over networks where each node updates its state by interacting with its neighbors and decides by itself whether certain level of agreement has been achieved among nodes. Since an individual node is unable to access the states of those beyond its neighbors, this problem...

This paper is concerned with a constrained optimization problem over a directed graph (digraph) of nodes, in which the cost function is a sum of local objectives, and each node only knows its local objective and constraints. To collaboratively solve the optimization, most of the existing works require the interaction graph to be balanced or "doubly...

This paper considers a distributed convex optimization problem with inequality constraints over time-varying unbalanced digraphs, where the cost function is a sum of local objectives, and each node of the graph only knows its local objective and inequality constraints. Although there is a vast literature on distributed optimization, most of them re...

This paper considers a distributed convex optimization problem with inequality constraints over time-varying unbalanced digraphs, where the cost function is a sum of local objectives, and each node of the graph only knows its local objective and inequality constraints. Although there is a vast literature on distributed optimization, most of them re...

Fabric dyeing is a critical production process in the clothing industry. Characterized by high energy consumption and water pollutant emission, dyeing processes need careful scheduling in order to reduce the relevant direct and indirect costs. In this paper, we describe the dyeing process scheduling problem as a bi-objective parallel batch-processi...

The industrial sector is one of the largest energy consumers in the world. To alleviate the grid’s burden during peak hours, time-of-use (TOU) electricity pricing has been implemented in many countries around the globe to encourage manufacturers to shift their electricity usage from peak periods to off-peak periods. In this paper, we study the unre...

This paper studies an option contract for coordinating a supply chain comprising one risk-neutral supplier and two risk-averse retailers engaged in promotion competition in the selling season. For a given option contract, in decentralized case, each risk-averse retailer decides the optimal order quantity and the promotion policy by maximizing the c...

Surrogate modelling based optimization has attracted much attention due to its ability of solving expensive-to-evaluate optimization problems, and a large majority of successful applications from various fields have been reported in literature. However, little effort has been devoted to solve scheduling problems through surrogate modelling, since e...

This study presents a novel approach to unsupervised learning for clustering with missing data. We first extend a finite mixture model to the infinite case by considering Dirichlet process mixtures, which can automatically determine the number of mixture components or clusters. Furthermore, we view the missing features as latent variables and compu...

Operational optimization of Chemical Mechanical Polishing, which sets the proper polishing time, is very important for improving the production efficiency of semiconductor manufacturing processes. However, usual operational optimization methods based on Run-to-Run strategies have not been suitable for the mixed-product processing mode of CMP. Also,...

The Affinity Propagation (AP) algorithm is an effective algorithm for clustering analysis, but it can not be directly applicable to the case of incomplete data. In view of the prevalence of missing data and the uncertainty of missing attributes, we put forward a modified AP clustering algorithm based on K -nearest neighbor intervals (KNNI) for inco...

This paper proposes new block properties for the flowshop scheduling problem with blocking to minimise makespan. A pruning procedure based on these proposed properties is used in the construction phase of an iterated greedy algorithm to decrease the total number of solutions to be examined to find an optimal schedule. Computational results using Ta...

Strong uncertainties is a key challenge for the application of scheduling algorithms in real-world production environments, since the optimized schedule at a time often turns to be deteriorated or even infeasible during its execution due to a large majority of unexpected events. This paper studies the uncertain scheduling problem arising from the s...

This paper is concerned with a binary detection problem over a non-secure network. To satisfy the communication rate constraint and against possible cyber attacks, which are modeled as deceptive signals injected to the network, a likelihood ratio based (LRB) scheduler is designed in the sensor side to smartly select sensor measurements for transmis...

Discriminative clustering is an unsupervised learning framework which introduces the discriminative learning rule of supervised classification into clustering. The underlying assumption is that a good partition (clustering) of the data should yield high discrimination, namely, the partitioned data can be easily classified by some classification alg...

This paper studies a periodic review inventory model with random supply capacity and demand, where the retailer is loss-averse. For the single-period problem, it is shown that the retailer will not order unless the initial inventory level is less than a critical value, and the order-up-to level is generally not a constant. Moreover, the critical va...

This paper proposes a Tabu-mechanism improved iterated greedy (TMIIG) algorithm to solve the no-wait flowshop scheduling problem with a makespan criterion. The idea of seeking further improvement in the iterated greedy (IG) algorithm framework is based on the observation that the construction phase of the original IG algorithm may not achieve good...

Recently, there has been an increasing concern on the carbon efficiency in the manufacturing industries. Since the carbon emissions in the industrial sector are directly related to the energy consumption for manufacturing, an effective way to improve carbon efficiency in an industrial plant is to design scheduling strategies to reduce the energy co...

We consider a three-dimensional problem of steering a nonholonomic vehicle to seek an unknown source of a spatially distributed signal field without any position measurement. In the literature, there exists an extremum seeking-based strategy under a constant forward velocity and tunable pitch and yaw velocities. Obviously, the vehicle with a consta...

For the minimisation of total tardiness in no-wait flowshops, objective incremental properties are investigated in this paper to speed up the evaluation of candidate solutions. To explore the properties, we introduce a new concept of sensitive jobs and identify through experiments that the proportion of such jobs is very small. Instead of evaluatin...

Extreme learning machines (ELMs) have proven to be efficient and effective learning mechanisms for pattern classification and regression. However, ELMs are primarily applied to supervised learning problems. Only a few existing research papers have used ELMs to explore unlabeled data. In this paper, we extend ELMs for both semi-supervised and unsupe...

In the operational optimization and scheduling problems of actual industrial processes, such as iron and steel, and microelectronics, the operational indices and process parameters usually need to be predicted. However, for some input and output variables of these prediction models, there may exist a lot of uncertainties coming from themselves, the...

Incomplete data are often encountered in data sets for clustering problems, and inappropriate treatment of incomplete data will significantly degrade the clustering performances. The Affinity Propagation (AP) algorithm is an effective algorithm for clustering analysis, but it is not directly applicable to the case of incomplete data. In view of the...

This paper proposes a tabu mechanism improved iterated greedy (TMIIG) algorithm to solve the no-wait flow-shop scheduling problem with makespan criterion. The motivation of seeking for further improvement in the iterated greedy (IG) algorithm framework is based on the observation that the construction phase of the original IG algorithm may lead to...

Many real-world networks exhibit overlapping community structure in which vertices may belong to more than one community. It has been recently shown that community structure plays an import role in epidemic spreading. However, the effect of different vertices on epidemic behavior was still unclear. In this paper, we classify vertices into overlappi...

In this paper a simple but efficient real-time detecting algorithm is proposed for tracking community structure of dynamic networks. Community structure is intuitively characterized as divisions of network nodes into subgroups, within which nodes are densely connected while between which they are sparsely connected. To evaluate the quality of commu...

This paper is concerned with a detection framework under scheduled communication for a binary hypothesis testing problem. A scheduler is designed to smartly select useful sensor measurements for transmission and leave non-useful ones, which results in that only a subset of measurements is sent to the testing agency. To this purpose, a likelihood ra...

In this paper, we focus on the relationship between operations-based variables (specifically, production speed, scrap rate and maintenance speed) and the manufacturing cost. These variables usually produce opposite influences on the variable cost and the fixed cost. For example, setting the production speed at a high level is beneficial for reducin...

Model structure selection is of crucial importance in radial basis function (RBF) neural networks. Existing model structure selection algorithms are essentially forward selection or backward elimination methods that may lead to sub-optimal models. This paper proposes an alternative selection procedure based on the kernelized least angle regression...

This paper studies a job shop scheduling problem with two new objective functions based on the setup and synergy costs besides the traditional total weighted tardiness criterion. The background is found in the real-world situation of a commercial vehicle producer, where the reduction of manufacturing costs has become a significant concern like in m...

This paper studies a single-period inventory problem with random yield and demand. In general, most of the previous works are based on the assumption of risk neutrality. We incorporate loss-averse preferences into this problem and the retailer's objective is to maximize the expected utility. We obtain the retailer's optimal ordering policy and then...

Long chain flexibility strategy is an effective way to match the supply with the uncertain demand in manufacturing system. However there are few studies on the long chain design problem with nonhomogeneous link costs. This paper first presents a mixed 0-1 LP model and proves that it belongs to NP-complete. Then an approximation algorithm is propose...

Due to the geological complexities of ore body formation and limited borehole sampling, this paper proposes a robust weighted least square support vector machine (LS-SVM) regression model to solve the ore grade estimation for a seafloor hydrothermal sulphide deposit in Solwara 1, which consists of a large proportion of incomplete samples without or...

Job shop scheduling is an important decision process in contemporary manufacturing systems. In this paper, we aim at the job shop scheduling problem in which the total weighted tardiness must be minimized. This objective function is relevant for the make-to-order production mode with an emphasis on customer satisfaction. In order to save the comput...

We consider a parallel machine scheduling problem with the objective of minimizing two types of costs: the cost related to production operations and the cost related to due date performances. The former could be reduced by reasonable settings of the operational variables (e.g., the number of workers, the frequency of maintenance), while the latter...

In this paper, we address the Bayesian classification with incomplete data. The common approach in the literature is to simply ignore the samples with missing values or impute missing values before classification. However, these methods are not effective when a large portion of the data have missing values and the acquisition of samples is expensiv...

This paper studies a single-period inventory problem with random yield and demand, where the loss-averse preferences are adopted to describe the retailer's (newsvendor's) decision-making behavior. When the loss-averse retailer orders, the fraction of good units in a batch is stochastic. He will choose an order quantity to maximize his expected util...

Large equipments are important research resources in colleges and universities, while the efficiently use into scientific research is a key issue. Few equipment management information systems (MIS) are built nowadays, even in areas like equipment reservation. Existing systems are geo-diverse and hard to be integrated, thus services that could enhan...

Currently the research of service science mainly focuses on the composition of distributed and heterogeneous Web services. Most of the approaches proposed to tackle this problem are based on cross-platform workflow and AI planning. However, until very recently, little attention is paid to the approaches based on complex networks. In the era of Big...

In real-world manufacturing systems, the processing of jobs is frequently affected by various unpredictable events. However, compared with the extensive research for the deterministic model, study on the random factors in job shop scheduling has not received sufficient attention. In this paper, we propose a hybrid differential evolution (DE) algori...

No-wait flowshop scheduling problem is widely investigated because of its practical application and specific properties. However, the total tardiness criterion has not been much considered. In this paper, we propose six heuristic approaches for no-wait flowshops with total tardiness criterion, among which the modified NEH algorithm (MNEH) is verifi...

In this paper, for permutation flowshops with two machines or more than two machines, as the number of jobs tends to infinity, the properties of asymptotic distribution of makespan are proposed. We introduce several conclusions in queuing theory to the scheduling problem, and convert the distribution of makespan to the distribution of waiting time...

The job shop scheduling problem (JSSP) has attracted much attention in the field of both information sciences and operations research. In terms of the objective function, most existing research has been focused on the makespan criterion (i.e., minimizing the overall completion time). However, for contemporary manufacturing firms, the due date relat...

This paper studies a newsvendor game in which two substitutable products are sold by two different retailers (newsvendors) with loss-averse preferences. Each loss-averse retailer facing stochastic customer demand and deterministic substitution rate will make an order quantity decision to maximize his expected utility. Since product substitution cau...

This paper considers a multi-period inventory problem with partially observed supply capacity in the lost sales case. Partially observed supply means that exact available supply in a period is observed only when the order quantity is not less than the supply capacity. Then, these observations are used to update the supply capacity distribution from...

The job shop scheduling problem (JSSP) is a notoriously difficult problem in combinatorial optimization. Up till now, neighbourhood search has been the most efficient optimization framework for solving the problem. The discovery and innovative application of neighbourhood properties is a key issue in the design of such algorithms. Effective neighbo...

This paper proposes a three-phase algorithm (TPA) for the flowshop scheduling problem with blocking (BFSP) to minimize makespan. In the first phase, the blocking nature of BFSP is exploited to develop a priority rule that creates a sequence of jobs. Using this as the initial sequence and a variant of the NEH-insert procedure, the second phase gener...

In this paper, a robust support vector regression (RSVR) method with uncertain input and output data is studied. First, the data uncertainties are investigated under a stochastic framework and two linear robust formulations are derived. Linear formulations robust to ellipsoidal uncertainties are also considered from a geometric perspective. Second,...

This paper proposes a novel constructive training algorithm for cascade neural networks. By reformulating the cascade neural network as a linear-in-the-parameters model, we use the orthogonal least squares (OLS) method to derive a novel objective function for training new hidden units. With this objective function, the sum of squared errors (SSE) o...