No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Quantifying the Impact of AI and Machine Learning on Data Access Optimization
ResearchGate has not been able to resolve any citations for this publication.
Reinforcement learning (RL) has become widely adopted in robot control. Despite many successes, one major persisting problem can be very low data efficiency. One solution is interactive feedback, which has been shown to speed up RL considerably. As a result, there is an abundance of different strategies, which are, however, primarily tested on discrete grid-world and small scale optimal control scenarios. In the literature, there is no consensus about which feedback frequency is optimal or at which time the feedback is most beneficial. To resolve these discrepancies we isolate and quantify the effect of feedback frequency in robotic tasks with continuous state and action spaces. The experiments encompass inverse kinematics learning for robotic manipulator arms of different complexity. We show that seemingly contradictory reported phenomena occur at different complexity levels. Furthermore, our results suggest that no single ideal feedback frequency exists. Rather that feedback frequency should be changed as the agent's proficiency in the task increases.
Machine learning has become a crucial tool in drug discovery and chemistry at large, e.g., to predict molecular properties, such as bioactivity, with high accuracy. However, activity cliffs─pairs of molecules that are highly similar in their structure but exhibit large differences in potency─have received limited attention for their effect on model performance. Not only are these edge cases informative for molecule discovery and optimization but also models that are well equipped to accurately predict the potency of activity cliffs have increased potential for prospective applications. Our work aims to fill the current knowledge gap on best-practice machine learning methods in the presence of activity cliffs. We benchmarked a total of 24 machine and deep learning approaches on curated bioactivity data from 30 macromolecular targets for their performance on activity cliff compounds. While all methods struggled in the presence of activity cliffs, machine learning approaches based on molecular descriptors outperformed more complex deep learning methods. Our findings highlight large case-by-case differences in performance, advocating for (a) the inclusion of dedicated "activity-cliff-centered" metrics during model development and evaluation and (b) the development of novel algorithms to better predict the properties of activity cliffs. To this end, the methods, metrics, and results of this study have been encapsulated into an open-access benchmarking platform named MoleculeACE (Activity Cliff Estimation, available on GitHub at: https://github.com/molML/MoleculeACE). MoleculeACE is designed to steer the community toward addressing the pressing but overlooked limitation of molecular machine learning models posed by activity cliffs.
Energy conservation and energy efficiency for smart antenna design is the reduction of energy consumption per unit of service or product in 5G Communication without reducing production quality and quantity. Efficient use of energy is important in many ways. First, fossil fuels such as oil and coal, which are important sources of energy, are depleting. Greenhouse gas emissions released into the atmosphere during energy production and consumption processes are major causes of climate change and global warming. In this paper, a smart energy storage and conservation model based on fuzzy logical controller was proposed to achieve maximum energy efficiency for smart antenna design in modern 5G Communication. For the initial level the proposed model regularly monitor the energy levels of different industrial components and then allot the energy as per the requirement of the components. If there any excess allocation required, then the proper requirement will request by the operator. Once the request is valid, the requirements will allocate to the components. The biggest factor that provides energy efficiency for smart antenna design is thermal insulation. Consuming less fuel means releasing less harmful gas into the atmosphere.
In today’s healthcare setting, the accurate and timely diagnosis of breast cancer is critical for recovery and treatment in the early stages. In recent years, the Internet of Things (IoT) has experienced a transformation that allows the analysis of real-time and historical data using artificial intelligence (AI) and machine learning (ML) approaches. Medical IoT combines medical devices and AI applications with healthcare infrastructure to support medical diagnostics. The current state-of-the-art approach fails to diagnose breast cancer in its initial period, resulting in the death of most women. As a result, medical professionals and researchers are faced with a tremendous problem in early breast cancer detection. We propose a medical IoT-based diagnostic system that competently identifies malignant and benign people in an IoT environment to resolve the difficulty of identifying early-stage breast cancer. The artificial neural network (ANN) and convolutional neural network (CNN) with hyperparameter optimization are used for malignant vs. benign classification, while the Support Vector Machine (SVM) and Multilayer Perceptron (MLP) were utilized as baseline classifiers for comparison. Hyperparameters are important for machine learning algorithms since they directly control the behaviors of training algorithms and have a significant effect on the performance of machine learning models. We employ a particle swarm optimization (PSO) feature selection approach to select more satisfactory features from the breast cancer dataset to enhance the classification performance using MLP and SVM, while grid-based search was used to find the best combination of the hyperparameters of the CNN and ANN models. The Wisconsin Diagnostic Breast Cancer (WDBC) dataset was used to test the proposed approach. The proposed model got a classification accuracy of 98.5% using CNN, and 99.2% using ANN.
Uncertainty quantification ( UQ ) is an important benchmark to assess the performance of artificial intelligence ( AI ) and particularly deep learning ensembled-based models. However, the ability for UQ using current AI -based methods is not only limited in terms of computational resources but it also requires changes to topology and optimization processes, as well as multiple performances to monitor model instabilities. From both geo-engineering and societal perspectives, a predictive groundwater table ( GWT ) model presents an important challenge, where a lack of UQ limits the validity of findings and may undermine science-based decisions. To overcome and address these limitations, a novel ensemble, an automated random deactivating connective weights approach ( ARDCW ), is presented and applied to retrieved geographical locations of GWT data from a geo-engineering project in Stockholm, Sweden. In this approach, the UQ was achieved via a combination of several derived ensembles from a fixed optimum topology subjected to randomly switched off weights, which allow predictability with one forward pass. The process was developed and programmed to provide trackable performance in a specific task and access to a wide variety of different internal characteristics and libraries. A comparison of performance with Monte Carlo dropout and quantile regression using computer vision and control task metrics showed significant progress in the ARDCW . This approach does not require changes in the optimization process and can be applied to already trained topologies in a way that outperforms other models.
The wire electric discharge machining (WEDM) is a potential alternative over the conventional machining methods, in terms of accuracy and ease in producing intricate shapes. However, the WEDM process parameters are exposed to unavoidable and unknown sources of uncertainties, following their inevitable influence over the process performance features. Thus, in the present work, we quantified the role of parametric uncertainty on the performance of the WEDM process. To this end, we used the practically relevant noisy experimental dataset to construct the four different machine learning (ML) models (linear regression, regression trees, support vector machines, and Gaussian process regression) and compared their goodness of fit based on the corresponding R ² and RMSE values. We further validated the prediction capability of the tested models by performing the error analysis. The model with the highest computational efficiency among the tested models is then used to perform data-driven uncertainty quantification and sensitivity analysis. The findings of the present article suggest that the pulse on time ( T on ) and peak current (IP) are the most sensitive parameters that influence the performance measures of the WEDM process. In this way, the current study achieves two goals: first, it proposes a predictive framework for determining the performance features of WEDM for unknown design points, and second, it reports data-driven uncertainty analysis in the light of parametric perturbations. The observations reported in the present article provide comprehensive computational insights into the performance characteristics of the WEDM process.
Artificial intelligence (AI) models based on deep learning now represent the state of the art for making functional predictions in genomics research. However, the underlying basis on which predictive models make such predictions is often unknown. For genomics researchers, this missing explanatory information would frequently be of greater value than the predictions themselves, as it can enable new insights into genetic processes. We review progress in the emerging area of explainable AI (xAI), a field with the potential to empower life science researchers to gain mechanistic insights into complex deep learning models. We discuss and categorize approaches for model interpretation, including an intuitive understanding of how each approach works and their underlying assumptions and limitations in the context of typical high-throughput biological datasets. In this Review, the authors describe advances in deep learning approaches in genomics, whereby researchers are moving beyond the typical ‘black box’ nature of models to obtain biological insights through explainable artificial intelligence (xAI).
Proton exchange membrane fuel cells (PEMFCs) as energy conversion devices for hydrogen energy are crucial for achieving an eco-friendly society, but their cost and performance are still not satisfactory for large-scale commercialization. Multiple physical and chemical coupling processes occur simultaneously at different scales in PEMFCs. Hence, previous studies only focused on the optimization of different components in such a complex system separately. In addition, the traditional trial-and-error method is very inefficient for achieving the performance breakthrough goal. Machine learning (ML) is a tool from the data science field. Trained based on datasets built from experimental records or theoretical simulation models, ML models can mine patterns that are difficult to draw intuitively. ML models can greatly reduce the cost of experimental attempts by predicting the target output. Serving as surrogate models, the ML approach could also greatly reduce the computational cost of numerical simulations such as first-principle or multiphysics simulations. Related reports are currently trending, and ML has been proven able to speed up tasks in this field, such as predicting active electrocatalysts, optimizing membrane electrode assembly (MEA), designing efficient flow channels, and providing stack operation strategies. Therefore, this paper reviews the applications and contributions of ML aiming at optimizing PEMFC performance regarding its potential to bring a research paradigm revolution. In addition to introducing and summarizing information for newcomers who are interested in this emerging cross-cutting field, we also look forward to and propose several directions for future development.
Machine learning (ML) models for predicting flotation behavior focus on operational variables. Fundamental aspects, e.g., physicochemical variables that describe mineral surfaces for bubble–particle interactions, are largely neglected in these models; however, these physicochemical variables of mineral particles, including bubbles and pulp, influence the flotation behavior. Thus, this study aimed to advance the prediction of flotation behavior by including physicochemical variables. Among four ML models used for the prediction, the random forest model had the best performance and was therefore subsequently used to investigate variable importance. Contact angle, particle diameter, bubble diameter, particle charge, collector concentration, flotation time, and number of mineral species were the most important variables. Limitations (e.g., assumptions and empiricism) and implications of our study were presented. Finally, our expectation was to encourage more attention to physicochemistry in flotation using ML for a more generalized empirical flotation model.
Deep learning technologies have achieved remarkable success in various tasks, ranging from computer vision, object detection to natural language processing. Unfortunately, state-of-the-art deep learning technologies are vulnerable to adversarial examples and backdoor attacks, where an adversary destroys the model’s integrity. The obstacles have urged intensive research on improving the ability of deep learning technologies to resist integrity attacks. However, existing defense methods are either incomplete (i.e., only a single attack can be detected) or expensive computing resources. It requires the defense method to have universal property, which can effectively and efficiently detect multiple integrity attacks. To this end, we propose a similarity-based integrity protection method for deep learning systems (IPDLS), which is provided with the universal property. IPDLS realizes anomaly detection by measuring the similarity between suspicious samples and samples in a preset verification set. We empirically evaluate IPDLS on the MNIST and CIFAR10 datasets. Experimental results have verified the effectiveness of IPDLS, which can detect adversarial examples and backdoor attacks simultaneously.
Nitrogen gas crossover (NGC) and nitrogen accumulation at the anode of proton exchange membrane (PEM) fuel cells are ineluctable and it would lead to inferior performance and even irreversible damage to functional components. To mitigate this issue, multiphysics numerical models (MNMs) are established to describe NGC behaviors and further guide experimental studies. However, to obtain the optimized parameters that would suppress NGC and retain high performance, grid search conducted on MSMs would cost unaffordable computational resources and time. Therefore, we innovatively introduced a machine learning-assisted MNM (MSM-ML) as a surrogate model, in which 9 state-of-the-art machine learning algorithms were compared, to greatly boost the resolution of this engineering problem. Through the proposed MSM-ML workflow performed on an experimentally validated MSM, the cost for obtaining the best parameter combination is greatly reduced. Moreover, the impact of each parameter in this complex system is directly revealed through the application of black-box interpretation methods afterwards. As a result, a new approach was pioneered to solve engineering problems which was demonstrated to be more efficient and intelligent than traditional methods. The NGC coefficient is reduced by 49.5%, while the power density is improved by 20% through the multivariable optimization of the developed MSM-ML.
Chloride diffusion coefficient is an important durability indicator in durability design of concrete structure according to performance-based approach. However, this indicator is difficult to anticipate and design due to a variety of factors such as mix design of concrete, the replacement of supplemental cementitious materials (SCMs), and binder selection. This study proposes an extensive number of machine learning algorithms to predict the chloride diffusion coefficient of concrete containing Supplementary Cementitious Materials (SCMs) such as silica fume, ground granulated blast furnace slag, and fly ash. A database containing nine input variables is created, eight machine learning models consisting of Support Vector Machine (SVM), Extreme Learning Machine (ELM), K-Nearest Neighbors (KNN), Light Gradient Boosting (LGB), Extreme Gradient Boosting (XGB), Random Forest (RF), Gradient Boosting (GB), AdaBoost (AdB) are evaluated via performance criteria such as Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE), coefficient of correlation ® and Root Mean Square Error (RMSE). Gradient Boosting model has highest performance in prediction of chloride diffusion coefficient. SHapley Additive exPlanations (SHAP), Individual Conditional Expectation ICE and Partial Dependence Plot PDP 2D allow the most influential inputs to be identified, quantify the influence of input variables on chloride diffusion of concrete. Selection of the best ML algorithm Gradient Boosting is useful to develop a dependable soft computing tool in durability design of concrete structure including the mix design optimization and binder selection.
In recent years, (big) data science has emerged as the “fourth paradigm” in physical science research. Data-driven techniques, e.g. machine learning, are advantageous in dealing with problems of high-dimensional features and complex mappings between quantities, which are otherwise of great difficulty or huge cost with other scientific paradigms. In the past five years or so, there has been a rapid growth of machine learning-assisted research on thermal transport. In this perspective, we review the recent progress in the intersection between machine learning and thermal transport, where machine learning methods generally serve as surrogate models for predicting the thermal transport properties, or as tools for designing structures for the desired thermal properties and exploring thermal transport mechanisms. We provide perspectives about the advantages of machine learning methods in comparison to the physics-based methods for studying thermal transport properties. We also discuss how to improve the accuracy of predictive analytics and efficiency of structural optimization, to provide guidance for better utilizing machine learning-based methods to advance thermal transport research. Finally, we identify several outstanding challenges in this active area as well as opportunities for future developments, including developing machine learning methods suitable for small datasets, discovering effective physics-based descriptors, generating dataset from experiments and validating machine learning results with experiments, and making breakthroughs via discovering new physics.
The quality of parts produced through laser powder bed fusion additive manufacturing can be irregular, with complex geometries sometimes exhibiting dimensional inaccuracies and defects. For optimal part quality, laser process parameters should be selected carefully prior to printing and adjusted during the print if necessary. This is challenging since approaches to control and optimize the build parameters need to take into account the part geometry, the material, and the complex physics of laser powder bed fusion. This work describes a data-driven approach using experimental diagnostics for the optimization of laser process parameters prior to printing. A training dataset is generated by collecting high speed photodiode signal data while printing simple parts containing key geometry features with various process parameter strategies. Supervised learning approaches are employed to train both a forward model and an inverse model. The forward model takes as inputs track-wise geometry features and laser parameters and outputs the photodiode signal along the scan path. The inverse model takes as inputs the geometry features and photodiode signal and predicts the laser parameters. Given the part geometry and a desired photodiode signal, the inverse model can thus determine the required laser parameters. Two test parts which contain defect-prone features are used to assess the validity of the inverse model. The use of the model leads to improved part quality (higher dimensional accuracy, reduced dross, reduced distortion) for both test geometries.
This article proposes a data-driven ticket dynamic pricing methodology for passenger railway service providers. There is a finite purchasing horizon, and the ticket prices should be set under varying conditions to affect the customer booking behaviour. A three-step process including machine learning and optimization tools is employed to maximize the revenue under a constrained train capacity. First, a multi-layer perceptron artificial neural network (MLP-ANN) model is proposed to predict the demand intensity due to seasonal situations using the ticket reservation data. Then, some regression models as price elasticity functions are used to quantify the effects of price, seasonal conditions and competition on the company’s sales. Finally, a nonlinear integer programming model is proposed to maximize the total revenue in the purchasing horizon. The results of the numerical studies on the Fadak Five-Star Trains’ reservation data indicate that the proposed methodology has high-grade potential to improve the service provider’s revenue.
Sustainable ai: Environmental implications, challenges and opportunities
- Wu
Sustainable ai: Environmental implications, challenges and opportunities
- C J Wu
- R Raghavendra
- U Gupta
- B Acun
- N Ardalani
- K Maeng