arXiv subjects

Hao Wang

Publications and source records attributed to Hao Wang.

At least 1,369 records · Page 76Linked to original sources

arXiv2022

IOHanalyzer: Detailed Performance Analyses for Iterative Optimization Heuristics

Benchmarking and performance analysis play an important role in understanding the behaviour of iterative optimization heuristics (IOHs) such as local search algorithms, genetic and evolutionary algorithms, Bayesian optimization algorithms, etc. This task, however, involves manual setup, execution, and analysis of the experiment on an individual basis, which is laborious and can be mitigated by a generic and well-designed platform. For this purpose, we propose IOHanalyzer, a new user-friendly tool for the analysis, comparison, and visualization of performance data of IOHs. Implemented in R and C++, IOHanalyzer is fully open source. It is available on CRAN and GitHub. IOHanalyzer provides detailed statistics about fixed-target running times and about fixed-budget performance of the benchmarked algorithms with a real-valued codomain, single-objective optimization tasks. Performance aggregation over several benchmark problems is possible, for example in the form of empirical cumulative distribution functions. Key advantages of IOHanalyzer over other performance analysis packages are its highly interactive design, which allows users to specify the performance measures, ranges, and granularity that are most useful for their experiments, and the possibility to analyze not only performance traces, but also the evolution of dynamic state parameters. IOHanalyzer can directly process performance data from the main benchmarking platforms, including the COCO platform, Nevergrad, the SOS platform, and IOHexperimenter. An R programming interface is provided for users preferring to have a finer control over the implemented functionalities.

cs.NE↗

arXiv2021

Automatic Configuration for Optimal Communication Scheduling in DNN Training

ByteScheduler partitions and rearranges tensor transmissions to improve the communication efficiency of distributed Deep Neural Network (DNN) training. The configuration of hyper-parameters (i.e., the partition size and the credit size) is critical to the effectiveness of partitioning and rearrangement. Currently, ByteScheduler adopts Bayesian Optimization (BO) to find the optimal configuration for the hyper-parameters beforehand. In practice, however, various runtime factors (e.g., worker node status and network conditions) change over time, making the statically-determined one-shot configuration result suboptimal for real-world DNN training. To address this problem, we present a real-time configuration method (called AutoByte) that automatically and timely searches the optimal hyper-parameters as the training systems dynamically change. AutoByte extends the ByteScheduler framework with a meta-network, which takes the system's runtime statistics as its input and outputs predictions for speedups under specific configurations. Evaluation results on various DNN models show that AutoByte can dynamically tune the hyper-parameters with low resource usage, and deliver up to 33.2\% higher performance than the best static configuration in ByteScheduler.

cs.DC↗

arXiv2021

Feature Selection for Efficient Local-to-Global Bayesian Network Structure Learning

Local-to-global learning approach plays an essential role in Bayesian network (BN) structure learning. Existing local-to-global learning algorithms first construct the skeleton of a DAG (directed acyclic graph) by learning the MB (Markov blanket) or PC (parents and children) of each variable in a data set, then orient edges in the skeleton. However, existing MB or PC learning methods are often computationally expensive especially with a large-sized BN, resulting in inefficient local-to-global learning algorithms. To tackle the problem, in this paper, we develop an efficient local-to-global learning approach using feature selection. Specifically, we first analyze the rationale of the well-known Minimum-Redundancy and Maximum-Relevance (MRMR) feature selection approach for learning a PC set of a variable. Based on the analysis, we propose an efficient F2SL (feature selection-based structure learning) approach to local-to-global BN structure learning. The F2SL approach first employs the MRMR approach to learn a DAG skeleton, then orients edges in the skeleton. Employing independence tests or score functions for orienting edges, we instantiate the F2SL approach into two new algorithms, F2SL-c (using independence tests) and F2SL-s (using score functions). Compared to the state-of-the-art local-to-global BN learning algorithms, the experiments validated that the proposed algorithms in this paper are more efficient and provide competitive structure learning quality than the compared algorithms.

cs.LG↗

arXiv2021

Time-of-use Pricing for Energy Storage Investment

Time-of-use (ToU) pricing is widely used by the electricity utility to shave peak load. Such a pricing scheme provides users with incentives to invest in behind-the-meter energy storage and to shift peak load towards low-price intervals. However, without considering the implication on energy storage investment, an improperly designed ToU pricing scheme may lead to significant welfare loss, especially when users over-invest the storage, which leads to new energy consumption peaks. In this paper, we will study how to design a social-optimum ToU pricing scheme by explicitly considering its impact on storage investment. We model the interactions between the utility and users as a two-stage optimization problem. To resolve the challenge of asymmetric information due to users' private storage cost, we propose a ToU pricing scheme based on different storage types and the aggregate demand per type. Each user does not need to reveal his private cost information. We can further compute the optimal ToU pricing with only a linear complexity. Simulations based on real-world data show that the suboptimality gap of our proposed ToU pricing, compared with the social optimum achieved under complete information, is less than 5%.

eess.SY↗

arXiv2021

Continual Learning with Knowledge Transfer for Sentiment Classification

This paper studies continual learning (CL) for sentiment classification (SC). In this setting, the CL system learns a sequence of SC tasks incrementally in a neural network, where each task builds a classifier to classify the sentiment of reviews of a particular product category or domain. Two natural questions are: Can the system transfer the knowledge learned in the past from the previous tasks to the new task to help it learn a better model for the new task? And, can old models for previous tasks be improved in the process as well? This paper proposes a novel technique called KAN to achieve these objectives. KAN can markedly improve the SC accuracy of both the new task and the old tasks via forward and backward knowledge transfer. The effectiveness of KAN is demonstrated through extensive experiments.

cs.CL↗

arXiv2021

Virtual Energy Storage Sharing and Capacity Allocation

Energy storage can play an important role in energy management of end users. To promote an efficient utilization of energy storage, we develop a novel business model to enable virtual storage sharing among a group of users. Specifically, a storage aggregator invests and operates the central physical storage unit, by virtualizing it into separable virtual capacities and selling to users. Each user purchases the virtual capacity, and utilize it to reduce the energy cost. We formulate the interaction between the aggregator and users as a two-stage optimization problem. In Stage 1, over the investment horizon, the aggregator determines the investment and pricing decisions. In Stage 2, in each operational horizon, each user decides the virtual capacity to purchase together with the operation of the virtual storage. We characterize a stepwise form of the optimal solution of Stage-2 Problem and a piecewise linear structure of the optimal profit of Stage-1 Problem, both with respect to the virtual capacity price. Based on the solution structure, we design an algorithm to attain the optimal solution of the two-stage problem. In our simulation results, the proposed storage virtualization model can reduce the physical energy storage investment of the aggregator by 54.3% and reduce the users' total costs by 34.7%, compared to the case where users acquire their own physical storage.

eess.SY↗

arXiv2021

Self-supervised Representation Learning Framework for Remote Physiological Measurement Using Spatiotemporal Augmentation Loss

Recent advances in supervised deep learning methods are enabling remote measurements of photoplethysmography-based physiological signals using facial videos. The performance of these supervised methods, however, are dependent on the availability of large labelled data. Contrastive learning as a self-supervised method has recently achieved state-of-the-art performances in learning representative data features by maximising mutual information between different augmented views. However, existing data augmentation techniques for contrastive learning are not designed to learn physiological signals from videos and often fail when there are complicated noise and subtle and periodic colour or shape variations between video frames. To address these problems, we present a novel self-supervised spatiotemporal learning framework for remote physiological signal representation learning, where there is a lack of labelled training data. Firstly, we propose a landmark-based spatial augmentation that splits the face into several informative parts based on the Shafer dichromatic reflection model to characterise subtle skin colour fluctuations. We also formulate a sparsity-based temporal augmentation exploiting Nyquist-Shannon sampling theorem to effectively capture periodic temporal changes by modelling physiological signal features. Furthermore, we introduce a constrained spatiotemporal loss which generates pseudo-labels for augmented video clips. It is used to regulate the training process and handle complicated noise. We evaluated our framework on 3 public datasets and demonstrated superior performances than other self-supervised methods and achieved competitive accuracy compared to the state-of-the-art supervised methods.

cs.CV↗

arXiv2021

Merging Subject Matter Expertise and Deep Convolutional Neural Network for State-Based Online Machine-Part Interaction Classification

Machine-part interaction classification is a key capability required by Cyber-Physical Systems (CPS), a pivotal enabler of Smart Manufacturing (SM). While previous relevant studies on the subject have primarily focused on time series classification, change point detection is equally important because it provides temporal information on changes in behavior of the machine. In this work, we address point detection and time series classification for machine-part interactions with a deep Convolutional Neural Network (CNN) based framework. The CNN in this framework utilizes a two-stage encoder-classifier structure for efficient feature representation and convenient deployment customization for CPS. Though data-driven, the design and optimization of the framework are Subject Matter Expertise (SME) guided. An SME defined Finite State Machine (FSM) is incorporated into the framework to prohibit intermittent misclassifications. In the case study, we implement the framework to perform machine-part interaction classification on a milling machine, and the performance is evaluated using a testing dataset and deployment simulations. The implementation achieved an average F1-Score of 0.946 across classes on the testing dataset and an average delay of 0.24 seconds on the deployment simulations.

cs.LG↗

arXiv2021

Electrocatalytic reduction of nitrogen to ammonia in ionic liquids

Ammonia NH3 is an important raw material for nitrogen fertilizer production and is widely used in industry. It is also a carbon free renewable fuel and an excellent hydrogen storage material. Reaction conditions of traditional Haber Bosch process are very harsh although it supports most of todays NH3 production. Recently, electrocatalytic nitrogen reduction reaction NRR has become one of the effective alternative methods to the traditional Haber Bosch process due to the advantages of mild operation conditions and almost no environmental pollutions. Ionic liquids ILs act as one of green solvents possess a lot of unique features and have shown excellent performance in electrocatalytic NRR process. In this review, we summarize fundamentals of the electrocatalytic NRR, the role of ILs in several typical electrocatalytic processes, and the possible role of ILs in the electrocatalytic NRR process. Moreover, the selection principles and issues of the application of ILs, as well as the future development direction for the electrocatalytic NRR process are concluded.

physics.chem-ph↗

arXiv2021

MatMat: Matrix Factorization by Matrix Fitting

Matrix factorization is a widely adopted recommender system technique that fits scalar rating values by dot products of user feature vectors and item feature vectors. However, the formulation of matrix factorization as a scalar fitting problem is not friendly to side information incorporation or multi-task learning. In this paper, we replace the scalar values of the user rating matrix by matrices, and fit the matrix values by matrix products of user feature matrix and item feature matrix. Our framework is friendly to multitask learning and side information incorporation. We use popularity data as side information in our paper in particular to enhance the performance of matrix factorization techniques. In the experiment section, we prove the competence of our method compared with other approaches using both accuracy and fairness metrics. Our framework is an ideal substitute for tensor factorization in context-aware recommendation and many other scenarios.

cs.IR↗

arXiv2021

Adaptive label thresholding methods for online multi-label classification

Existing online multi-label classification works cannot well handle the online label thresholding problem and lack the regret analysis for their online algorithms. This paper proposes a novel framework of adaptive label thresholding algorithms for online multi-label classification, with the aim to overcome the drawbacks of existing methods. The key feature of our framework is that both scoring and thresholding models are included as important components of the online multi-label classifier and are incorporated into one online optimization problem. Further, in order to establish the relationship between scoring and thresholding models, a novel multi-label classification loss function is derived, which measures to what an extent the multi-label classifier can distinguish between relevant labels and irrelevant ones for an incoming instance. Based on this new framework and loss function, we present a first-order linear algorithm and a second-order one, which both enjoy closed form update, but rely on different techniques for updating the multi-label classifier. Both algorithms are proved to achieve a sub-linear regret. Using Mercer kernels, our first-order algorithm has been extended to deal with nonlinear multi-label prediction tasks. Experiments show the advantage of our linear and nonlinear algorithms, in terms of various multi-label performance metrics.

cs.LG↗

arXiv2021

ProtGNN: Towards Self-Explaining Graph Neural Networks

Despite the recent progress in Graph Neural Networks (GNNs), it remains challenging to explain the predictions made by GNNs. Existing explanation methods mainly focus on post-hoc explanations where another explanatory model is employed to provide explanations for a trained GNN. The fact that post-hoc methods fail to reveal the original reasoning process of GNNs raises the need of building GNNs with built-in interpretability. In this work, we propose Prototype Graph Neural Network (ProtGNN), which combines prototype learning with GNNs and provides a new perspective on the explanations of GNNs. In ProtGNN, the explanations are naturally derived from the case-based reasoning process and are actually used during classification. The prediction of ProtGNN is obtained by comparing the inputs to a few learned prototypes in the latent space. Furthermore, for better interpretability and higher efficiency, a novel conditional subgraph sampling module is incorporated to indicate which part of the input graph is most similar to each prototype in ProtGNN+. Finally, we evaluate our method on a wide range of datasets and perform concrete case studies. Extensive results show that ProtGNN and ProtGNN+ can provide inherent interpretability while achieving accuracy on par with the non-interpretable counterparts.

cs.LG↗

arXiv2021

Anapole-Mediated Emission Enhancement in Gallium Nitride Nanocavities

Benefiting from their low-loss light manipulation at subwavelength scales, optically resonant dielectric nanostructures have emerged as one of the most promising nanophotonic building blocks. Here, we theoretically conceive a dielectric nanocavity made of moderate-refractive-index gallium nitride and investigate the strong electromagnetic field confinement inside the nanocavity. We demonstrate that gallium nitride nanodisks can support anapole states, which result from interference between electric dipole and toroidal dipole modes and are tunable by changing sizes of the nanodisks. The highly confined electromagnetic field of the anapole states can promote the emission efficiency of a single quantum emitter inside the nanocavity. Moreover, the emission polarization can be tuned by placing the quantum emitter off the nanodisk center. Our findings provide a promising candidate for the construction of ultra-compact, super-radiative integrated quantum light sources.

physics.optics↗

arXiv2021

Effect of a weak magnetic field on ductile-brittle transition in micro-cutting of single-crystal calcium fluoride

Magneto-plasticity occurs when a weak magnetic field alters material plasticity and offers a viable solution to enhance ductile-mode cutting of brittle materials. This study demonstrates the susceptibility of non-magnetic single-crystal calcium fluoride (CaF2) to the magneto-plastic effect. The influence of magneto-plasticity on CaF2 was confirmed in micro-deformation tests under a weak magnetic field of 20 mT. The surface pile-up effect was weakened by 10-15 nm along with an enlarged plastic zone and suppressed crack propagation under the influence of the magnetic field. Micro-cutting tests along different crystal orientations on the (111) plane of CaF2 revealed an increase in the ductile-brittle transition of the machined surface with the aid of magneto-plasticity where the largest increase in ductile-brittle transition occurred along the [11-2] orientation from 512 nm to a range of 664-806 nm. Meanwhile, the subsurface damage layer was concurrently thinner under magnetic influence. An anisotropic influence of the magnetic field relative to the single-crystal orientation and the cutting direction was also observed. An analytical model was derived to determine an orientation factor M that successfully describes the anisotropy while considering the single-crystal dislocation behaviour, material fracture toughness, and the orientation of the magnetic field. Previously suggested theoretical mechanism of magneto-plasticity via formation of non-singlet electronic states in defected configurations was confirmed with density functional theory calculations. The successful findings on the influence of a weak magnetic field on plasticity present an opportunity for the adoption of magnetic-assisted micro-cutting of non-magnetic materials.

cond-mat.mtrl-sci↗

arXiv2021

Data privacy protection in microscopic image analysis for material data mining

Recent progress in material data mining has been driven by high-capacity models trained on large datasets. However, collecting experimental data has been extremely costly owing to the amount of human effort and expertise required. Therefore, material researchers are often reluctant to easily disclose their private data, which leads to the problem of data island, and it is difficult to collect a large amount of data to train high-quality models. In this study, a material microstructure image feature extraction algorithm FedTransfer based on data privacy protection is proposed. The core contributions are as follows: 1) the federated learning algorithm is introduced into the polycrystalline microstructure image segmentation task to make full use of different user data to carry out machine learning, break the data island and improve the model generalization ability under the condition of ensuring the privacy and security of user data; 2) A data sharing strategy based on style transfer is proposed. By sharing style information of images that is not urgent for user confidentiality, it can reduce the performance penalty caused by the distribution difference of data among different users.

eess.IV↗

arXiv2021

Machine Learning Approach to Uncovering Residential Energy Consumption Patterns Based on Socioeconomic and Smart Meter Data

The smart meter data analysis contributes to better planning and operations for the power system. This study aims to identify the drivers of residential energy consumption patterns from the socioeconomic perspective based on the consumption and demographic data using machine learning. We model consumption patterns by representative loads and reveal the relationship between load patterns and socioeconomic characteristics. Specifically, we analyze the real-world smart meter data and extract load patterns by clustering in a robust way. We further identify the influencing socioeconomic attributes on load patterns to improve our method's interpretability. The relationship between consumers' load patterns and selected socioeconomic features is characterized via machine learning models. The findings are as follows. (1) Twelve load clusters, consisting of six for weekdays and six for weekends, exhibit a diverse pattern of lifestyle and a difference between weekdays and weekends. (2) Among various socioeconomic features, age and education level are suggested to influence the load patterns. (3) Our proposed analytical model using feature selection and machine learning is proved to be more effective than XGBoost and conventional neural network model in mapping the relationship between load patterns and socioeconomic features.

cs.LG↗

arXiv2021

Dynamic control of speed and trajectories of active droplets in a nematic environment by electric field and focused laser beam

One objective of active matter science is to unveil principles by which chaotic microscale dynamics could be transformed into useful work. A nematic liquid crystal environment offers a number of possibilities, one of which is a directional motion of an active droplet filled with an aqueous dispersion of swimming bacteria. In this work, using the responsiveness of the nematic to the electric field and light, we demonstrate how to control the direction and speed of active droplets. The dielectric response of nematic to the electric field causes two effects: (i) reorientation of the overall director, and (ii) changing the symmetry of the director configuration around the droplet. The first effect redirects the propulsion direction while the second one changes the speed. A laser beam pointed to the vicinity of the droplet can trigger the desired director symmetry around the droplet, by switching between dipolar and quadrupolar configurations, thus affecting the motility and polarity of propulsion. The dynamic tuning of the direction and speed of active droplets represents a step forward in the development of controllable microswimmers.

cond-mat.soft↗

arXiv2021

Path-Enhanced Multi-Relational Question Answering with Knowledge Graph Embeddings

The multi-relational Knowledge Base Question Answering (KBQA) system performs multi-hop reasoning over the knowledge graph (KG) to achieve the answer. Recent approaches attempt to introduce the knowledge graph embedding (KGE) technique to handle the KG incompleteness but only consider the triple facts and neglect the significant semantic correlation between paths and multi-relational questions. In this paper, we propose a Path and Knowledge Embedding-Enhanced multi-relational Question Answering model (PKEEQA), which leverages multi-hop paths between entities in the KG to evaluate the ambipolar correlation between a path embedding and a multi-relational question embedding via a customizable path representation mechanism, benefiting for achieving more accurate answers from the perspective of both the triple facts and the extra paths. Experimental results illustrate that PKEEQA improves KBQA models' performance for multi-relational question answering with explainability to some extent derived from paths.

cs.CL↗