2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)最新文献

{"title":"MATGAN: Unified GANs for Multimodal Attribute Transfer by Coarse-to-Fine Disentangling Representations","authors":"Xi Guo, Qiang Rao, Kun He, Fang Chen, Bing Yu, Bailan Feng, Jian Huang, Qin Yang","doi":"10.1109/FAIML57028.2022.00030","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00030","url":null,"abstract":"Image attribute transfer aims to change an image to a target one with desired attributes. There are mainly two challenges for this task: multi-domain transfer and attribute-level multimodality. The first means editing multiple attributes using a single model and the second means diverse appearances for the target attribute. Existing methods cannot address the two problems simultaneously. Moreover, many works focus on image-level multimodality rather than attribute-level. In this paper, we propose a novel coarse-to-fine disentangling representation framework MATGAN to achieve Multimodal Attribute Transfer. In the coarse disentangling stage, we propose to embed images onto a content space and an attribute space for image-level multimodality. In the fine disentangling stage, we further disentangle the attribute space to bind with each attribute for attribute-level multimodal and multi-domain transfer. Extensive experiments demonstrate the effectiveness of our approach.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121967954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Statistical Analysis of Connected and Autonomous Vehicles (CAVs) Effects on the Environment in Terms of Pollutants and Fuel Consumption","authors":"Alireza Ansariyar, Safieh Laaly","doi":"10.1109/FAIML57028.2022.00037","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00037","url":null,"abstract":"Over the last decades, interest in integrating autonomous and connected technologies in vehicle design in order to improve mobility, safety, and reduce transportation's environmental impact has dramatically increased. The state-of-the-art specified that connected and autonomous vehicles (CAVs) ameliorate traffic mobility, safety, fuel/energy consumption, and reduce environmental pollution. The State of Maryland (MD) in the United States was selected as a case study, and the paper appraised CAVs' fuel consumption and air pollutants (CO, PM, and NOx), and utilized reasonable linear regression models to forecast CAV's environmental effects. The VISUM software was applied to simulate MD transport network as a multi-modal transport network and the required data on a set of variables were collected through an exhaustive survey. The amount of pollutants and fuel consumption were obtained for timestamps 2010 to 2021 from the macro simulation. Eventually, four linear regression models were suggested to predict the amount of CO, NOx, PM pollutants and, fuel consumption. The results demonstrated that CAVs' pollutants and fuel consumption have a significant correlation with income, age, and race of the CAV customers. Moreover, the reliability of four statistical models was compared with the reliability of macro simulation model outputs in year 2030. The error values of three pollutants and fuel consumption were obtained less than 9% by statistical models in SPSS. This research is expected to assist researchers and policymakers with planning decisions to reduce CAV environmental impacts in MD.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115454691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tawa Pukllay Proof: New Method for Solving Arithmetic Operations with The Inca Yupana Using Pattern Recognition and Parallelism","authors":"Dhavit Prem, Rosario Guzman-Jimenez, Fernando Sotomayor, Alvaro Saldivar","doi":"10.1109/FAIML57028.2022.00048","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00048","url":null,"abstract":"Yupana is an Inca device used for arithmetic operations. This article describes a new arithmetical system: Tawa Pukllay (TP), where arithmetic operations do not require mental calculations: no carries, no borrows, no memorization of multiplication tables, nor trial and error procedures for divisions. Instead, user recognizes patterns and makes predefined movements to perform the four basic arithmetic operations very quickly; moreover, the result of the operation can be reached by multiple paths and in parallel, allowing each user to create his own strategies. This paper proves with mathematical rigor that TP produces correct numerical results.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116625796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SARAH: Semantic-Aware Representation Balance Hashing for Image Retrieval","authors":"Changlin Fan, Fengming Liang, Bo Xiao, Yuqiong Wu, Jincheng Yu, Shifei Zhou, Ye Li, Chunjie Sheng","doi":"10.1109/FAIML57028.2022.00039","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00039","url":null,"abstract":"Deep hashing is vitally important for large-scale image retrieval. Recently, central similarity based deep hashing approaches have shown great advantages for category-level image retrieval; in the existing approaches, however, categories are typically represented by a set of predefined binary vectors which are generated from Hadamard matrix or entry-wisely sampled from Bernoulli distribution. Unfortunately, such kind of category representations lack of discriminativity and semantic information. In this paper, we propose a novel Semantic-Aware Representation bAlance Hashing framework, dubbed SARAH, for category-level image retrieval. Specifically, in SARAH, the category representations are learned to preserve semantic similarities and to maximize pairwise distance; whereas the continuous code of each image is extracted by convolutional network and supervised via a central similarity loss with the corresponding semantic representation which is constructed by the learned category representations. As a consequence, the semantically similar images can be encoded to hash codes with small Hamming distance.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129570127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Defining the Settings of Product Attributes for Product Design Using an Innovative NSGA-II","authors":"Huimin Jiang, Farzad Sabetzadeh","doi":"10.1109/FAIML57028.2022.00029","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00029","url":null,"abstract":"In the process of product design, the quality function deployment and affective design are needed in the initial stage. In fact, for a new product, affective design needs designers, and engineering design needs engineers to complete. Generally, some design attributes in affective design and engineering specifications in engineering design are common. However, the two processes, quality function deployment and affective design, are carried out separately, so there will be some differences in the settings of the same design attributes and engineering specifications. This paper formulates a multi-objective optimization model which considers the above two processes. In order to optimize the design, an innovative non-dominated sorting genetic algorithm-II (NSGA-II) based on chaos optimization algorithm (COA) is proposed. At the same time, using the way of case analysis, the mobile phone is selected as the research object to illustrate the establishment of the model and the optimization technique.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130864655","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Program Committee","authors":"Mihaela A. Bornea","doi":"10.1109/ASE.2006.55","DOIUrl":"https://doi.org/10.1109/ASE.2006.55","url":null,"abstract":"RDFox is a new materialisation-based RDF system currently being developed at Oxford University. The system is currently RAM-based, and its algorithms have been designed to take full advantage of modern multi-core/processor systems. In my talk I will present an overview of some of the techniques we developed in the context of the RDFox project. In particular, I will discuss our algorithm that parallelises computation with very little overhead, I will present an overview of our lock-free indexes for RDF data, and I will discuss a novel incremental update algorithm. I will also briefly talk about some issues that we are currently working on, such as improving query planning and distributing data in a cluster of servers. The NPD Benchmark for OBDA Systems Davide Lanti, Martin Rezk, Mindaugas Slusnys, Guohui Xiao, and Diego Calvanese Faculty of Computer Science, Free University of Bozen-Bolzano, Italy Abstract. In Ontology-Based Data Access (OBDA), queries are posed over a high-level conceptual view, and then translated into queries over a potentially very large (usually relational) data source. The ontology is connected to the data sources through a declarative specification given in terms of mappings. Although prototype OBDA systems providing the ability to answer SPARQL queries over the ontology are available, a significant challenge remains: performance. To properly evaluate OBDA systems, benchmarks tailored towards the requirements in this setting are needed. OWL benchmarks, which have been developed to test the performance of generic SPARQL query engines, however, fail to evaluate OBDA specific features. In this work, we propose a novel benchmark for OBDA systems based on the Norwegian Petroleum Directorate (NPD). Our benchmark comes with novel techniques to generate, from available data, datasets of increasing size, taking into account the requirements dictated by the OBDA setting. We validate our benchmark on significant OBDA systems, showing that it is more adequate than previous benchmarks not tailored for OBDA. In Ontology-Based Data Access (OBDA), queries are posed over a high-level conceptual view, and then translated into queries over a potentially very large (usually relational) data source. The ontology is connected to the data sources through a declarative specification given in terms of mappings. Although prototype OBDA systems providing the ability to answer SPARQL queries over the ontology are available, a significant challenge remains: performance. To properly evaluate OBDA systems, benchmarks tailored towards the requirements in this setting are needed. OWL benchmarks, which have been developed to test the performance of generic SPARQL query engines, however, fail to evaluate OBDA specific features. In this work, we propose a novel benchmark for OBDA systems based on the Norwegian Petroleum Directorate (NPD). Our benchmark comes with novel techniques to generate, from available data, datasets of increasing size, taking into accou","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117264339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on the Application of Hotel Cleanliness Compliance Detection Algorithm Based on WGAN","authors":"Xiang Kang, Hui Gao","doi":"10.1109/FAIML57028.2022.00027","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00027","url":null,"abstract":"Aiming at the problems of irregular cleaning and supervision difficulties in the cleaning process of hotel bathrooms, a target detection algorithm based on deep learning is proposed to detect the cleaning process transmitted by the sensor in real time and analyze its prescriptivity. However, the cleaning process has factors such as occlusion, light influence and insufficient data volume, resulting in inefficient detection. Therefore, this paper proposes a deep convolutional generation adversarial network (DCGAN) as the basic framework to expand the data set, improve the adaptability and robustness of the detector to different detection targets, take advantage of the fast speed and high accuracy of the YOLOv5 target detection network to detect the target, and then design a compliance detection network algorithm to detect whether the target meets the cleanliness standards. Experimental results show that the method has rapidity, practicality and high accuracy, and fully meets the engineering needs of hotel cleaning process detection and supervision.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"181 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131718917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Function of Traffic Prediction in Alleviating Traffic Congestion","authors":"Zheng Zhao, Zhenxing Han, Changchen Zhao, Yixin Zhang","doi":"10.1109/FAIML57028.2022.00035","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00035","url":null,"abstract":"Traffic prediction technique can be used to guide people's travel, and there are many researches have been conducted to achieve a higher prediction accuracy. If the traffic prediction result can be timely conveyed to all travelers, personal travel plan may also change accordingly, thus to influence the traffic state of the road network. This paper considers decision-making model, cyclic causal model, and game process, and analyzes the function of traffic prediction in alleviating traffic congestion. Simulation results prove that the sharing mechanism of traffic forecast results is very important, meanwhile, a limited effect can be also found when using traffic prediction to alleviate traffic congestion.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116428969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Manufacturing Feature Recognition Method Based on Subgraph Decomposition","authors":"Jingning Wu, Ruoshan Lei, Yibing Peng","doi":"10.1109/FAIML57028.2022.00031","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00031","url":null,"abstract":"Manufacturing feature recognition is a key technology to realize the integration of design and manufacturing. This paper firstly proposes a subgraph decomposition method based on the mechanical part representation form of Generalized Extended Attribute Adjacency Graph (GEAAG), which decomposes the overall mechanical part graph into the multiple subgraphs. Secondly, based on the predefined manufacturing feature Attribute Adjacency Graph (AAG) and extended attribute rules, the sub-atlas are traversed to match the real manufacturing features to realize the recognition of manufacturing features. Finally, a prototype system was developed to validate the proposed method, and a comparative experiment with the traditional method is conducted to verify that the presented method has higher recognition efficiency.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130262871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Applications of a Braess Paradox Traffic Management Software","authors":"P. Joseph","doi":"10.1109/FAIML57028.2022.00036","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00036","url":null,"abstract":"The Traffic Management Software System developed in this study is a vast improvement upon existing systems, as it makes use of Braess' paradox to individualize and optimize the routes drivers take to their destination. Braess' paradox states that, as drivers tend to make selfish decisions regarding their path, drivers will all elect to take any faster, more efficient path opened - thus increasing travel time on that path, and our traffic management software system makes use of this paradox by individualizing routes for drivers so that they do not all take the same shortcut or, in the event of construction or an accident, the same detour, thereby clogging it. This is an improvement on existing traffic management systems because existing traffic management systems will direct all drivers to the same route, which increases the volume of traffic on these routes and the amount of time it takes to travel on them. If there is construction or a car accident on a given road, all cars will be directed to the same detour route, which will create a high volume of traffic on that road. If a new, shorter road is opened, drivers will all be directed to use it, and the road will become clogged. These roads are always suggested to drivers no matter what the traffic volume is. However, our traffic management software disincentivizes, or renders temporarily unusable, high - traffic roads due to the high amount of congestion. However, this problem is solved with the customization of routes for individual drivers, and the opening and closing of certain routes to drivers based on their traffic volume. Routes are output by the software system by using the demand and capacity of these routes, and the travel time on them, to generate 'Braess routes', which are routes deemed efficient by the software and a function of route demand and travel time. When a Braess route becomes congested, it is disincentivized, thus eventually eliminating the high traffic. This is achieved using the Frank Wolfe algorithm, which formulates and minimizes linear approximations of the route demand and travel time functions used to output the Braess routes. The use of these individualized Braess routes that don't direct all drivers to the same shortcut or the same detour, and the disincentivization of congested routes both help to reduce road congestion and travel time. A SUMO GUI is specifically a visual interface that was able to be implemented when models of the traffic system were present. This indicates that our software was able to correctly identify the Braess routes as they continually changed, as well as that our original hypothesis, that applying Braess' paradox to a traffic management software and customizing routes to maximize efficiency for individual drivers would decrease overall travel time. Therefore, we have developed a system that can apply the Frank-Wolfe Algorithm and Braess' paradox in order to identify new or changing Braess routes. Our findings have also shown that the use","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130047989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}