Yao Wan, Zhangqian Bi, Yang He, Jianguo Zhang, Hongyu Zhang, Yulei Sui, Guandong Xu, Hai Jin, Philip Yu
Code intelligence leverages machine learning techniques to extract knowledge from extensive code corpora, with the aim of developing intelligent tools to improve the quality and productivity of computer programming. Currently, there is already a thriving research community focusing on code intelligence, with efforts ranging from software engineering, machine learning, data mining, natural language processing, and programming languages. In this paper, we conduct a comprehensive literature review on deep learning for code intelligence, from the aspects of code representation learning, deep learning techniques, and application tasks. We also benchmark several state-of-the-art neural models for code intelligence, and provide an open-source toolkit tailored for the rapid prototyping of deep-learning-based code intelligence models. In particular, we inspect the existing code intelligence models under the basis of code representation learning, and provide a comprehensive overview to enhance comprehension of the present state of code intelligence. Furthermore, we publicly release the source code and data resources to provide the community with a ready-to-use benchmark, which can facilitate the evaluation and comparison of existing and future code intelligence models (https://xcodemind.github.io). At last, we also point out several challenging and promising directions for future research.
{"title":"Deep Learning for Code Intelligence: Survey, Benchmark and Toolkit","authors":"Yao Wan, Zhangqian Bi, Yang He, Jianguo Zhang, Hongyu Zhang, Yulei Sui, Guandong Xu, Hai Jin, Philip Yu","doi":"10.1145/3664597","DOIUrl":"https://doi.org/10.1145/3664597","url":null,"abstract":"<p>Code intelligence leverages machine learning techniques to extract knowledge from extensive code corpora, with the aim of developing intelligent tools to improve the quality and productivity of computer programming. Currently, there is already a thriving research community focusing on code intelligence, with efforts ranging from software engineering, machine learning, data mining, natural language processing, and programming languages. In this paper, we conduct a comprehensive literature review on deep learning for code intelligence, from the aspects of code representation learning, deep learning techniques, and application tasks. We also benchmark several state-of-the-art neural models for code intelligence, and provide an open-source toolkit tailored for the rapid prototyping of deep-learning-based code intelligence models. In particular, we inspect the existing code intelligence models under the basis of code representation learning, and provide a comprehensive overview to enhance comprehension of the present state of code intelligence. Furthermore, we publicly release the source code and data resources to provide the community with a ready-to-use benchmark, which can facilitate the evaluation and comparison of existing and future code intelligence models (https://xcodemind.github.io). At last, we also point out several challenging and promising directions for future research.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141251603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Self-tuning is a feature of autonomic databases that includes the problem of automatic schema design. It aims at providing an optimized schema that increases the overall database performance. While in relational databases automatic schema design focuses on the automated design of the physical schema, in NoSQL databases all levels of representation are considered: conceptual, logical, and physical. This is mainly because the latter are mostly schema-less and lack a standard schema design procedure as is the case for SQL databases. In this work, we carry out a systematic literature survey on automatic schema design in both SQL and NoSQL databases. We identify the levels of representation and the methods that are used for the schema design problem, and we present a novel taxonomy to classify and compare different schema design solutions. Our comprehensive analysis demonstrates that, despite substantial progress that has been made, schema design is still a developing field and considerable challenges need to be addressed, notably for NoSQL databases. We highlight the most important findings from the results of our analysis and identify areas for future research work.
{"title":"Self-tuning Database Systems: A Systematic Literature Review of Automatic Database Schema Design and Tuning","authors":"M. Mozaffari, Anton Dignös, J. Gamper, U. Störl","doi":"10.1145/3665323","DOIUrl":"https://doi.org/10.1145/3665323","url":null,"abstract":"\u0000 Self-tuning is a feature of autonomic databases that includes the problem of automatic schema design. It aims at providing an optimized schema that increases the overall database performance. While in relational databases automatic schema design focuses on the automated design of the physical schema, in NoSQL databases all levels of representation are considered: conceptual, logical, and physical. This is mainly because the latter are mostly schema-less and lack a standard schema design procedure as is the case for SQL databases. In this work, we carry out a systematic literature survey on automatic schema design in both SQL\u0000 and\u0000 NoSQL databases. We identify the levels of representation and the methods that are used for the schema design problem, and we present a novel taxonomy to classify and compare different schema design solutions. Our comprehensive analysis demonstrates that, despite substantial progress that has been made, schema design is still a developing field and considerable challenges need to be addressed, notably for NoSQL databases. We highlight the most important findings from the results of our analysis and identify areas for future research work.\u0000","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140963290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shaoxiong Ji, Xiaobo Li, Wei Sun, Hang Dong, Ara Taalas, Yijia Zhang, Honghan Wu, Esa Pitkänen, Pekka Marttinen
Automated medical coding, an essential task for healthcare operation and delivery, makes unstructured data manageable by predicting medical codes from clinical documents. Recent advances in deep learning and natural language processing have been widely applied to this task. However, deep learning-based medical coding lacks a unified view of the design of neural network architectures. This review proposes a unified framework to provide a general understanding of the building blocks of medical coding models and summarizes recent advanced models under the proposed framework. Our unified framework decomposes medical coding into four main components, i.e., encoder modules for text feature extraction, mechanisms for building deep encoder architectures, decoder modules for transforming hidden representations into medical codes, and the usage of auxiliary information. Finally, we introduce the benchmarks and real-world usage and discuss key research challenges and future directions.
{"title":"A Unified Review of Deep Learning for Automated Medical Coding","authors":"Shaoxiong Ji, Xiaobo Li, Wei Sun, Hang Dong, Ara Taalas, Yijia Zhang, Honghan Wu, Esa Pitkänen, Pekka Marttinen","doi":"10.1145/3664615","DOIUrl":"https://doi.org/10.1145/3664615","url":null,"abstract":"<p>Automated medical coding, an essential task for healthcare operation and delivery, makes unstructured data manageable by predicting medical codes from clinical documents. Recent advances in deep learning and natural language processing have been widely applied to this task. However, deep learning-based medical coding lacks a unified view of the design of neural network architectures. This review proposes a unified framework to provide a general understanding of the building blocks of medical coding models and summarizes recent advanced models under the proposed framework. Our unified framework decomposes medical coding into four main components, i.e., encoder modules for text feature extraction, mechanisms for building deep encoder architectures, decoder modules for transforming hidden representations into medical codes, and the usage of auxiliary information. Finally, we introduce the benchmarks and real-world usage and discuss key research challenges and future directions.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141251600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The metaverse delivered through converged and amalgamated technologies holds promise. No wonder technology heavyweights, large corporates, research organizations and businesses cutting across industry verticals are racing to put in place a metaverse-first strategy. The bets on consumers rapidly migrating from traditional social networks and collaborative applications to more immersive digital experiences have been placed. However, the transition is not expected to be seamless. Privacy, safety and security concerns abound in the early versions of the metaverse. Increased regulatory oversight and diverse national laws threaten to derail the hype around the metaverse. It is increasingly clear that the final iteration of the metaverse will need to assuage the concerns of individual users while addressing complex legal and regulatory requirements. Thus, a multi-perspective approach needs to be adopted to help set the agenda for the evolution of the metaverse. This research paper examines the different aspects and challenges which the future metaverse will need to address. A set of ”first principles” are formulated, which if implemented will lead to the development of an equitable, inclusive, safe and secure metaverse.
{"title":"The First Principles: Setting the Context for a Safe and Secure Metaverse","authors":"Ankur Gupta, Sahil Sawhney, Kashyap Kompella","doi":"10.1145/3665495","DOIUrl":"https://doi.org/10.1145/3665495","url":null,"abstract":"The metaverse delivered through converged and amalgamated technologies holds promise. No wonder technology heavyweights, large corporates, research organizations and businesses cutting across industry verticals are racing to put in place a metaverse-first strategy. The bets on consumers rapidly migrating from traditional social networks and collaborative applications to more immersive digital experiences have been placed. However, the transition is not expected to be seamless. Privacy, safety and security concerns abound in the early versions of the metaverse. Increased regulatory oversight and diverse national laws threaten to derail the hype around the metaverse. It is increasingly clear that the final iteration of the metaverse will need to assuage the concerns of individual users while addressing complex legal and regulatory requirements. Thus, a multi-perspective approach needs to be adopted to help set the agenda for the evolution of the metaverse. This research paper examines the different aspects and challenges which the future metaverse will need to address. A set of ”first principles” are formulated, which if implemented will lead to the development of an equitable, inclusive, safe and secure metaverse.","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140963893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The field of Virtual Reality continues to evolve to provide an ever-greater sense of immersion to the user. However, VR experiences are still primarily constrained through the human senses of vision and audition, with some interest in haptic (mainly vibrotactile) applications. Only recently have olfactory displays- technologies that generate and deliver scent stimuli- been examined to provide the sense of smell to the human olfactory organ in virtual environments. This paper presents a classification and review of olfactory-enhanced virtual reality systems, particularly those that deployed a Head-Mounted Display (HMD) or Cave Automatic Virtual Environment (CAVE) system. In addition, the paper provides a discussion of the various technological and design challenges for developing an olfactory display suitable for enhancing virtual reality experiences. Finally, the paper proposes future perspectives on the field and includes a table summarizing the characteristics and features of the reviewed systems.
{"title":"A Review of Olfactory Display Designs for Virtual Reality Environments","authors":"Jordan Tewell, Nimesha Ranasinghe","doi":"10.1145/3665243","DOIUrl":"https://doi.org/10.1145/3665243","url":null,"abstract":"The field of Virtual Reality continues to evolve to provide an ever-greater sense of immersion to the user. However, VR experiences are still primarily constrained through the human senses of vision and audition, with some interest in haptic (mainly vibrotactile) applications. Only recently have olfactory displays- technologies that generate and deliver scent stimuli- been examined to provide the sense of smell to the human olfactory organ in virtual environments. This paper presents a classification and review of olfactory-enhanced virtual reality systems, particularly those that deployed a Head-Mounted Display (HMD) or Cave Automatic Virtual Environment (CAVE) system. In addition, the paper provides a discussion of the various technological and design challenges for developing an olfactory display suitable for enhancing virtual reality experiences. Finally, the paper proposes future perspectives on the field and includes a table summarizing the characteristics and features of the reviewed systems.","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140972505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shangsong Liang, Zhou Pan, wei liu, Jian Yin, M. de Rijke
Recommender systems have become an important instrument to connect people to information. Sparse, complex, and rapidly growing data presents new challenges to traditional recommendation algorithms. To overcome these challenges, various deep learning-based recommendation algorithms have been proposed. Among these, Variational AutoEncoder (VAE)-based recommendation methods stand out. VAEs are based on a flexible probabilistic framework, which is robust for data sparsity and compatible with other deep learning-based models for dealing with multimodal data. In addition, the deep generative structure of VAEs helps to perform Bayesian inference in an efficient manner. VAE-based recommendation algorithms have given rise to many novel graphical models and they have achieved promising performance. In this paper, we conduct a survey to systematically summarize recent VAE-based recommendation algorithms. Four frequently used characteristics of VAE-based recommendation algorithms are summarized, and a taxonomy of VAE-based recommendation algorithms is proposed. We also identify future research directions for, advanced perspectives on, and the application of VAEs in recommendation algorithms, to inspire future work on VAEs for recommender systems.
{"title":"A Survey on Variational Autoencoders in Recommender Systems","authors":"Shangsong Liang, Zhou Pan, wei liu, Jian Yin, M. de Rijke","doi":"10.1145/3663364","DOIUrl":"https://doi.org/10.1145/3663364","url":null,"abstract":"Recommender systems have become an important instrument to connect people to information. Sparse, complex, and rapidly growing data presents new challenges to traditional recommendation algorithms. To overcome these challenges, various deep learning-based recommendation algorithms have been proposed. Among these, Variational AutoEncoder (VAE)-based recommendation methods stand out. VAEs are based on a flexible probabilistic framework, which is robust for data sparsity and compatible with other deep learning-based models for dealing with multimodal data. In addition, the deep generative structure of VAEs helps to perform Bayesian inference in an efficient manner. VAE-based recommendation algorithms have given rise to many novel graphical models and they have achieved promising performance. In this paper, we conduct a survey to systematically summarize recent VAE-based recommendation algorithms. Four frequently used characteristics of VAE-based recommendation algorithms are summarized, and a taxonomy of VAE-based recommendation algorithms is proposed. We also identify future research directions for, advanced perspectives on, and the application of VAEs in recommendation algorithms, to inspire future work on VAEs for recommender systems.","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140972006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alexandre Heuillet, Ahmad Nasser, Hichem Arioui, Hedi Tabia
In the past few years, Differentiable Neural Architecture Search (DNAS) rapidly imposed itself as the trending approach to automate the discovery of deep neural network architectures. This rise is mainly due to the popularity of DARTS (Differentiable ARchitecTure Search), one of the first major DNAS methods. In contrast with previous works based on Reinforcement Learning or Evolutionary Algorithms, DNAS is faster by several orders of magnitude and uses fewer computational resources. In this comprehensive survey, we focused specifically on DNAS and reviewed recent approaches in this field. Furthermore, we proposed a novel challenge-based taxonomy to classify DNAS methods. We also discussed the contributions brought to DNAS in the past few years and its impact on the global NAS field. Finally, we concluded by giving some insights into future research directions for the DNAS field.
{"title":"Efficient Automation of Neural Network Design: A Survey on Differentiable Neural Architecture Search","authors":"Alexandre Heuillet, Ahmad Nasser, Hichem Arioui, Hedi Tabia","doi":"10.1145/3665138","DOIUrl":"https://doi.org/10.1145/3665138","url":null,"abstract":"<p>In the past few years, Differentiable Neural Architecture Search (DNAS) rapidly imposed itself as the trending approach to automate the discovery of deep neural network architectures. This rise is mainly due to the popularity of DARTS (Differentiable ARchitecTure Search), one of the first major DNAS methods. In contrast with previous works based on Reinforcement Learning or Evolutionary Algorithms, DNAS is faster by several orders of magnitude and uses fewer computational resources. In this comprehensive survey, we focused specifically on DNAS and reviewed recent approaches in this field. Furthermore, we proposed a novel challenge-based taxonomy to classify DNAS methods. We also discussed the contributions brought to DNAS in the past few years and its impact on the global NAS field. Finally, we concluded by giving some insights into future research directions for the DNAS field.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141251577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pietro Melzi, Christian Rathgeb, Ruben Tolosana, Ruben Vera, Christoph Busch
Privacy-enhancing technologies are technologies that implement fundamental data protection principles. With respect to biometric recognition, different types of privacy-enhancing technologies have been introduced for protecting stored biometric data which are generally classified as sensitive. In this regard, various taxonomies and conceptual categorizations have been proposed and standardisation activities have been carried out. However, these efforts have mainly been devoted to certain sub-categories of privacy-enhancing technologies and therefore lack generalization. This work provides an overview of concepts of privacy-enhancing technologies for biometric recognition in a unified framework. Key properties and differences between existing concepts are highlighted in detail at each processing step. Fundamental characteristics and limitations of existing technologies are discussed and related to data protection techniques and principles. Moreover, scenarios and methods for the assessment of privacy-enhancing technologies for biometric recognition are presented. This paper is meant as a point of entry to the field of data protection for biometric recognition applications and is directed towards experienced researchers as well as non-experts.
{"title":"An Overview of Privacy-Enhancing Technologies in Biometric Recognition","authors":"Pietro Melzi, Christian Rathgeb, Ruben Tolosana, Ruben Vera, Christoph Busch","doi":"10.1145/3664596","DOIUrl":"https://doi.org/10.1145/3664596","url":null,"abstract":"<p>Privacy-enhancing technologies are technologies that implement fundamental data protection principles. With respect to biometric recognition, different types of privacy-enhancing technologies have been introduced for protecting stored biometric data which are generally classified as sensitive. In this regard, various taxonomies and conceptual categorizations have been proposed and standardisation activities have been carried out. However, these efforts have mainly been devoted to certain sub-categories of privacy-enhancing technologies and therefore lack generalization. This work provides an overview of concepts of privacy-enhancing technologies for biometric recognition in a unified framework. Key properties and differences between existing concepts are highlighted in detail at each processing step. Fundamental characteristics and limitations of existing technologies are discussed and related to data protection techniques and principles. Moreover, scenarios and methods for the assessment of privacy-enhancing technologies for biometric recognition are presented. This paper is meant as a point of entry to the field of data protection for biometric recognition applications and is directed towards experienced researchers as well as non-experts.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141256840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
jiaxu leng, Yongming Ye, Mengjingcheng MO, Chenqiang Gao, Ji Gan, Bin Xiao, Xinbo Gao
Aerial object detection, as object detection in aerial images captured from an overhead perspective, has been widely applied in urban management, industrial inspection, and other aspects. However, the performance of existing aerial object detection algorithms is hindered by variations in object scales and orientations attributed to the aerial perspective. This survey presents a comprehensive review of recent advances in aerial object detection. We start with some basic concepts of aerial object detection and then summarize the five imbalance problems of aerial object detection, including scale imbalance, spatial imbalance, objective imbalance, semantic imbalance, and class imbalance. Moreover, we classify and analyze relevant methods and especially introduce the applications of aerial object detection in practical scenarios. Finally, the performance evaluation is presented on two popular aerial object detection datasets VisDrone-DET and DOTA, and we discuss several future directions that could facilitate the development of aerial object detection.
{"title":"Recent Advances for Aerial Object Detection: A Survey","authors":"jiaxu leng, Yongming Ye, Mengjingcheng MO, Chenqiang Gao, Ji Gan, Bin Xiao, Xinbo Gao","doi":"10.1145/3664598","DOIUrl":"https://doi.org/10.1145/3664598","url":null,"abstract":"<p>Aerial object detection, as object detection in aerial images captured from an overhead perspective, has been widely applied in urban management, industrial inspection, and other aspects. However, the performance of existing aerial object detection algorithms is hindered by variations in object scales and orientations attributed to the aerial perspective. This survey presents a comprehensive review of recent advances in aerial object detection. We start with some basic concepts of aerial object detection and then summarize the five imbalance problems of aerial object detection, including scale imbalance, spatial imbalance, objective imbalance, semantic imbalance, and class imbalance. Moreover, we classify and analyze relevant methods and especially introduce the applications of aerial object detection in practical scenarios. Finally, the performance evaluation is presented on two popular aerial object detection datasets VisDrone-DET and DOTA, and we discuss several future directions that could facilitate the development of aerial object detection.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140915139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carlos Núñez-Molina, Pablo Mesejo, Juan Fernández-Olivares
In the field of Sequential Decision Making (SDM), two paradigms have historically vied for supremacy: Automated Planning (AP) and Reinforcement Learning (RL). In the spirit of reconciliation, this paper reviews AP, RL and hybrid methods (e.g., novel learn to plan techniques) for solving Sequential Decision Processes (SDPs), focusing on their knowledge representation: symbolic, subsymbolic or a combination. Additionally, it also covers methods for learning the SDP structure. Finally, we compare the advantages and drawbacks of the existing methods and conclude that neurosymbolic AI poses a promising approach for SDM, since it combines AP and RL with a hybrid knowledge representation.
在顺序决策(SDM)领域,历来有两种范式争夺主导地位:自动规划(AP)和强化学习(RL)。本着和解的精神,本文回顾了用于解决序列决策过程(SDP)的自动规划、强化学习和混合方法(如新颖的学习规划技术),重点是它们的知识表示:符号表示、次符号表示或组合表示。此外,它还包括学习 SDP 结构的方法。最后,我们比较了现有方法的优缺点,并得出结论:神经符号人工智能是一种很有前途的 SDM 方法,因为它将 AP 和 RL 与混合知识表示法相结合。
{"title":"A Review of Symbolic, Subsymbolic and Hybrid Methods for Sequential Decision Making","authors":"Carlos Núñez-Molina, Pablo Mesejo, Juan Fernández-Olivares","doi":"10.1145/3663366","DOIUrl":"https://doi.org/10.1145/3663366","url":null,"abstract":"<p>In the field of Sequential Decision Making (SDM), two paradigms have historically vied for supremacy: Automated Planning (AP) and Reinforcement Learning (RL). In the spirit of reconciliation, this paper reviews AP, RL and hybrid methods (e.g., novel learn to plan techniques) for solving Sequential Decision Processes (SDPs), focusing on their knowledge representation: symbolic, subsymbolic or a combination. Additionally, it also covers methods for learning the SDP structure. Finally, we compare the advantages and drawbacks of the existing methods and conclude that neurosymbolic AI poses a promising approach for SDM, since it combines AP and RL with a hybrid knowledge representation.</p>","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":null,"pages":null},"PeriodicalIF":16.6,"publicationDate":"2024-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140907232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}