3,070 Hits in 9.3 sec

Autonomous Extraction of a Hierarchical Structure of Tasks in Reinforcement Learning, A Sequential Associate Rule Mining Approach [article]

Behzad Ghazanfari, Fatemeh Afghah, Matthew E. Taylor
2018 arXiv   pre-print
In this paper, we propose a novel method based on Sequential Association Rule Mining that can extract Hierarchical Structure of Tasks in Reinforcement Learning (SARM-HSTRL) in an autonomous manner for  ...  The proposed method leverages association rule mining to discover the causal and temporal relationships among states in different trajectories, and extracts a task hierarchy that captures these relationships  ...  Conclusion A HRL method called SARM-HSTRL is proposed to autonomously extract a task hierarchy for RL by utilizing a sequential associate role mining approach, where multiple subgoals are extracted as  ... 
arXiv:1811.08275v1 fatcat:xdcl7yfh6rfupmloa6vryimp2e

Autonomous Extracting a Hierarchical Structure of Tasks in Reinforcement Learning and Multi-task Reinforcement Learning [article]

Behzad Ghazanfari, Matthew E. Taylor
2017 arXiv   pre-print
We introduce a novel method called ARM-HSTRL (Association Rule Mining to extract Hierarchical Structure of Tasks in Reinforcement Learning).  ...  This paper proposes a novel practical method that can autonomously decompose tasks, by leveraging association rule mining, which discovers hidden relationship among entities in data mining.  ...  In fact, the combination of HST and ARM is a sequential association rule mining procedure.  ... 
arXiv:1709.04579v2 fatcat:pj3g6f5zvnaxlkmyo65bdyznc4

L-Diversity Based Dynamic Update for Large Time-Evolving Microdata [chapter]

Xiaoxun Sun, Hua Wang, Jiuyong Li
2008 Lecture Notes in Computer Science  
We go further and show how a problem can be automatically decomposed into a partial-order task-hierarchy, and solved using hierarchical reinforcement learning.  ...  Association mining often produces large collections of association rules that are difficult to understand and put into action.  ... 
doi:10.1007/978-3-540-89378-3_47 fatcat:3srytfhtszejdidu62nvzwia3m

Explainable Artificial Intelligence (xAI) Approaches and Deep Meta-Learning Models [chapter]

Evren Dağlarli
2020 Advances and Applications in Deep Learning  
In classical artificial intelligence approaches, we frequently encounter deep learning methods available today.  ...  These deep learning methods can yield highly effective results according to the data set size, data set quality, the methods used in feature extraction, the hyper parameter set used in deep learning models  ...  Therefore, this approach requires a hierarchical structure that learns to learn a new task with distributed hierarchically structured metadata.  ... 
doi:10.5772/intechopen.92172 fatcat:sgmxtwloa5bbzb5sp7tpi75i3y

Web mining in soft computing framework: relevance, state of the art and future directions

S.K. Pal, V. Talwar, P. Mitra
2002 IEEE Transactions on Neural Networks  
A survey of the existing literature on "soft web mining" is provided along with the commercially available systems.  ...  The reason for considering web mining, a separate field from data mining, is explained.  ...  The WEBMINER [24] system proposes a structured query language (SQL)-like querying mechanism for querying the discovered knowledge (in the form of association rules and sequential patterns).  ... 
doi:10.1109/tnn.2002.1031947 pmid:18244512 fatcat:a2ea5nfnczgjlpwsbwe6ebt5hi

Knowledge Augmented Machine Learning with Applications in Autonomous Driving: A Survey [article]

Julian Wörmann, Daniel Bogdoll, Etienne Bührle, Han Chen, Evaristus Fuh Chuo, Kostadin Cvejoski, Ludger van Elst, Tobias Gleißner, Philip Gottschall, Stefan Griesche, Christian Hellert, Christian Hesels (+34 others)
2022 arXiv   pre-print
The identified approaches are structured according to the categories integration, extraction and conformity. Special attention is given to applications in the field of autonomous driving.  ...  The existence of representative datasets is a prerequisite of many successful artificial intelligence and machine learning models.  ...  Because with this Association Rule Mining approach thousands of rules would be generated without pruning the input, it takes a lot of space and iterations to mine those rules.  ... 
arXiv:2205.04712v1 fatcat:u2bgxr2ctnfdjcdbruzrtjwot4

Sequential Pattern Mining for Situation and Behavior Prediction in Simulated Robotic Soccer [chapter]

Andreas D. Lattner, Andrea Miene, Ubbo Visser, Otthein Herzog
2006 Lecture Notes in Computer Science  
In this work we present an approach which applies unsupervised symbolic learning off-line to a qualitative abstraction in order to create frequent patterns in dynamic scenes.  ...  In order to allow agents to act autonomously and to make their decisions on a solid basis an interpretation of the current scene is necessary.  ...  Acknowledgment The work in this paper was partially funded by the Deutsche Forschungsgemeinschaft (DFG) in the SPP-1125.  ... 
doi:10.1007/11780519_11 fatcat:2sosommn6bcc7ninlz7lit2iqi

Machine Learning and Data Mining Methods in Diabetes Research

Ioannis Kavakiotis, Olga Tsave, Athanasios Salifoglou, Nicos Maglaveras, Ioannis Vlahavas, Ioanna Chouvarda
2017 Computational and Structural Biotechnology Journal  
In general, 85% of those used were characterized by supervised learning approaches and 15% by unsupervised ones, and more specifically, association rules.  ...  The aim of the present study is to conduct a systematic review of the applications of machine learning, data mining techniques and tools in the field of diabetes research with respect to a) Prediction  ...  Acknowledgements This work has been partially supported by Horizon 2020 Framework Programme of the European Union under grant agreement 644906, the AEGLE project.  ... 
doi:10.1016/j.csbj.2016.12.005 pmid:28138367 pmcid:PMC5257026 fatcat:gq3lcg5i7jal7ps45vrbje6ufu

Exploiting the knowledge engineering paradigms for designing smart learning systems

Abdel Badeeh Mohamed M. Salem, Silvia Parusheva
2018 Eastern-European Journal of Enterprise Technologies  
Salem, Silvia Parusheva, 2018 Literature review and problem statement Many of KE and intelligent algorithms are used in the developing of smart learning systems, e.g. artificial neural networks, support  ...  This section reviews some of the most important aspects of knowledge engineering  ...  In a particular case, we may be interested in the determination of the link between the variables. f) Sequence Analysis: this type of task is oriented to problems of modeling sequential data.  ... 
doi:10.15587/1729-4061.2018.128410 fatcat:wnpj7lsg4ffstjdjqfyezyesqu

Intrinsically Motivated Open-Ended Multi-Task Learning Using Transfer Learning to Discover Task Hierarchy

Nicolas Duminy, Sao Mai Nguyen, Junshuai Zhu, Dominique Duhaut, Jerome Kerdreux
2021 Applied Sciences  
In open-ended continuous environments, robots need to learn multiple parameterised control tasks in hierarchical reinforcement learning.  ...  We show with a simulation and a real industrial robot arm, in cross-task and cross-learner transfer settings, that task composition is key to tackle highly complex tasks.  ...  The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.  ... 
doi:10.3390/app11030975 fatcat:4inw4fj5ajgnzf3nm6x5aghhvu

The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors [article]

William H. Guss, Cayden Codel, Katja Hofmann, Brandon Houghton, Noboru Kuno, Stephanie Milani, Sharada Mohanty, Diego Perez Liebana, Ruslan Salakhutdinov, Nicholay Topin, Manuela Veloso, Phillip Wang
2021 arXiv   pre-print
To that end, we introduce: (1) the Minecraft ObtainDiamond task, a sequential decision making environment requiring long-term planning, hierarchical control, and efficient exploration methods; and (2)  ...  Though deep reinforcement learning has led to breakthroughs in many difficult domains, these successes have required an ever-increasing number of samples.  ...  We request poster stands or materials for hanging the posters of the Round 2 participants.  ... 
arXiv:1904.10079v3 fatcat:n3xfk2fyfnhe7l5oafdmccfbza


2004 Advances in Fuzzy Systems — Applications and Theory  
What are the most important problems of computational intelligence? A sketch of the road to intelligent systems is presented.  ...  Acknowledgments We would like to thank our expert colleagues who supported this project by sending descriptions of problems that according to them are the most challenging issues in the field of computational  ...  This is important for many data mining tasks, such as classification, clustering, and rule extraction.  ... 
doi:10.1142/9789812562531_0001 fatcat:lk3kdecperbt3aynsbd7jnqsoe

Multiagent Systems for Network Intrusion Detection: A Review [chapter]

Álvaro Herrero, Emilio Corchado
2009 Advances in Intelligent and Soft Computing  
So far, plenty of techniques have been applied for the detection of intrusions, which has been reported in many surveys.  ...  This work focuses the development of network-based IDSs from an architectural point of view, in which multiagent systems are applied for the development of IDSs, presenting an up-to-date revision of the  ...  When applied to the ID problem, an association-rules algorithm determines the relationships between the different fields in audit trails, while a meta-learning classifier learns the signatures of attacks  ... 
doi:10.1007/978-3-642-04091-7_18 dblp:conf/cisis-spain/HerreroC09 fatcat:qo6lklarvvgihkqwz2iqr7iofy

ABLE: A toolkit for building multiagent autonomic systems

J. P. Bigus, D. A. Schlosnagle, J. R. Pilgrim, W. N. Mills III, Y. Diao
2002 IBM Systems Journal  
Our approach to building autonomic systems is based on combining autonomous intelligent agents in a well-structured way.  ...  Data and learning AbleBeans are combined to create neural classification, neural clustering, and neural prediction agents that can be used for lightweight data mining tasks.  ...  Mills helps build ABLE-based applications for use as embedded "rules engines" to help analyze and mine data, to make real-time assessments of various application states to drive problem diagnosis, and  ... 
doi:10.1147/sj.413.0350 fatcat:shavvhlc6fd25jtdzmacolwx3y

The MineRL 2020 Competition on Sample Efficient Reinforcement Learning using Human Priors [article]

William H. Guss, Mario Ynocente Castro, Sam Devlin, Brandon Houghton, Noboru Sean Kuno, Crissman Loomis, Stephanie Milani, Sharada Mohanty, Keisuke Nakata, Ruslan Salakhutdinov, John Schulman, Shinya Shiroshita (+3 others)
2021 arXiv   pre-print
Although deep reinforcement learning has led to breakthroughs in many difficult domains, these successes have required an ever-increasing number of samples, affording only a shrinking segment of the AI  ...  To that end, participants compete under a limited environment sample-complexity budget to develop systems which solve the MineRL ObtainDiamond task in Minecraft, a sequential decision making environment  ...  expressive power as a simulator to make great strides in language-grounded, interpretable multi-task option-extraction, hierarchical lifelong learning, and active perception.  ... 
arXiv:2101.11071v1 fatcat:gzd6vohfavaypnz2vqey6tgkqa
« Previous Showing results 1 — 15 out of 3,070 results