Filters








21,007 Hits in 3.3 sec

Hierarchical Representation Learning for Markov Decision Processes [article]

Lorenzo Steccanella, Simone Totaro, Anders Jonsson
2021 arXiv   pre-print
In this paper we present a novel method for learning hierarchical representations of Markov decision processes.  ...  We empirically validate the method, by showing that it can successfully learn a useful hierarchical representation in a navigation domain.  ...  Background In this section we define Markov decision processes and hierarchical reinforcement learning in the form of the options framework.  ... 
arXiv:2106.01655v2 fatcat:bzje2xuahreg3f255rk2zgaagu

Detecting and Responding to Concept Drift in Business Processes

Lingkai Yang, Sally McClean, Mark Donnelly, Kevin Burke, Kashaf Khan
2022 Algorithms  
In this paper, we model a response to concept drift as a sequential decision making problem by combing a hierarchical Markov model and a Markov decision process (MDP).  ...  Concept drift, which refers to changes in the underlying process structure or customer behaviour over time, is inevitable in business processes, causing challenges in ensuring that the learned model is  ...  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/a15050174 fatcat:3n4gnl33dfeatok6r6ofvzg42a

Object-Oriented Representation and Hierarchical Reinforcement Learning in Infinite Mario

M. Joshi, R. Khobragade, S. Sarda, U. Deshpande, S. Mohan
2012 2012 IEEE 24th International Conference on Tools with Artificial Intelligence  
We then use this representation combined with the hierarchical reinforcement learning model as a learning framework.  ...  With the help of experimental results, we show that this approach facilitates faster and efficient learning for this domain. (Abstract)  ...  Semi-Markov Decision Process The semi-Markov decision process (SMDP) is a generalization of the MDP which incorporates temporally extended actions.  ... 
doi:10.1109/ictai.2012.152 dblp:conf/ictai/JoshiKSDM12 fatcat:dkj4fj6jonb57oxcsvtkwb67qy

Higher-Order Logic [chapter]

John Lloyd
2017 Encyclopedia of Machine Learning and Data Mining  
Starting from the concept of regular Markov models we introduce the concept of hidden Markov model, and the issue of estimating the output emission and transition probabilities between hidden states, for  ...  We mention typical application in which hidden Markov models play a central role, and mention a number of popular implementations.  ...  Semi-Markov Decision Problem Formalism The common underlying formalism in hierarchical reinforcement learning is the semi-Markov decision process (SMDP).  ... 
doi:10.1007/978-1-4899-7687-1_126 fatcat:oya3su4vhjh3lk77kbsw2qe2y4

A Hierarchical Approach For The Design Of Gesture-To-Sound Mappings

Jules Françoise, Baptiste Caramiaux, Frédéric Bevilacqua
2012 Proceedings of the SMC Conferences  
We thank all members of the Real Time Musical Interactions team and members of the Legos project for fruitful discussions.  ...  Similarly, Jordà [19, 20] argued for the need of considering different control levels allowing for either intuitive or compositional decisions.  ...  GESTURE SEGMENTATION USING HIERARCHICAL MARKOV MODELS We explain in this section how Hierarchical Hidden Markov Models can be used in our framework.  ... 
doi:10.5281/zenodo.850029 fatcat:vwwtrmarpremvhuozfssd42qb4

Learning agents for uncertain environments (extended abstract)

Stuart Russell
1998 Proceedings of the eleventh annual conference on Computational learning theory - COLT' 98  
This seems to be a very interesting problem for the COLT, UAI, and ML communities, and has been addressed in econometrics under the heading of structural estimation of Markov decision processes.  ...  I will discuss methods for learning the parameters and structure of such representations from sensory inputs, and for computing posterior probabilities.  ...  Reinforcement learning (RL) methods are essentially online algorithmd for solving Markov decision processes (MDPs).  ... 
doi:10.1145/279943.279964 dblp:conf/colt/Russell98 fatcat:bnv35r7dzfdzfdhch3qea6amdy

:{unav)

Andrew G. Barto, Sridhar Mahadevan
2012 Discrete event dynamic systems  
Common to these approaches is a reliance on the theory of semi-Markov decision processes, which we emphasize in our review.  ...  We then discuss extensions of these ideas to concurrent activities, multiagent coordination, and hierarchical memory for addressing partial observability.  ...  We follow Sutton and Barto (1998) in denoting the reward for the action at stage t by r t 1 instead of the more usual r t . 2.  ... 
doi:10.1023/a:1025696116075 fatcat:l7go63ebc5abrf2je75k5h6fau

Structured Prediction of Sequences and Trees Using Infinite Contexts [chapter]

Ehsan Shareghi, Gholamreza Haffari, Trevor Cohn, Ann Nicholson
2015 Lecture Notes in Computer Science  
To facilitate learning of this large and unbounded model, we use a hierarchical Pitman-Yor process prior which provides a recursive form of smoothing.  ...  We propose a novel hierarchical model for structured prediction over sequences and trees which exploits global context by conditioning each generation decision on an unbounded context of prior decisions  ...  To facilitate learning of such a large and unbounded model, the predictive distributions associated with tree contexts are smoothed in a recursive manner using a hierarchical Pitman-Yor process.  ... 
doi:10.1007/978-3-319-23525-7_23 fatcat:kpevaszn65a3xnsbyblskvvlk4

Structured Prediction of Sequences and Trees using Infinite Contexts [article]

Ehsan Shareghi, Gholamreza Haffari, Trevor Cohn, Ann Nicholson
2015 arXiv   pre-print
To facilitate learning of this large and unbounded model, we use a hierarchical Pitman-Yor process prior which provides a recursive form of smoothing.  ...  We propose a novel hierarchical model for structured prediction over sequences and trees which exploits global context by conditioning each generation decision on an unbounded context of prior decisions  ...  To facilitate learning of such a large and unbounded model, the predictive distributions associated with tree contexts are smoothed in a recursive manner using a hierarchical Pitman-Yor process.  ... 
arXiv:1503.02417v1 fatcat:7txfrppwkvc5tnutoteisub6ue

Dimensions of complexity of intelligent agents

David Poole, Alan Mackworth
2006 Proceedings of the 2006 international symposium on Practical cognitive agents and robots - PCAR '06  
Research has progressed by making simplifying assumptions about the representations of the agents or about the environments the agents act in.  ...  For each of these dimensions, there is a simplified case and progressively more complex cases.  ...  observable Markov decision processes GT Game theory representations of strategic and extensive form of a game.  ... 
doi:10.1145/1232425.1232438 fatcat:rda4bue57ndy3frw2trmbu5fjq

Assisted living technologies for older adults

Parisa Rashidi
2012 Proceedings of the 2nd ACM SIGHIT symposium on International health informatics - IHI '12  
Decision process (MDP)  Opportunity Knocks (OK): public transit assistance [Patterson 2004]  iRoute: Learns walking preference of dementia patients [Hossain 2011]  Commercial b ■ h d ■ o p ■ ■  ...  Task # Robots Social Communication 46 Hobbies 29 New Learning 16 Data from Understanding the potential for robot assistance for older adults in the home environment (HFA-TR-1102).  ... 
doi:10.1145/2110363.2110478 dblp:conf/ihi/Rashidi12 fatcat:vavobpvbqzfslm4343duxh7yfe

Representational efficiency outweighs action efficiency in human program induction [article]

Sophia Sanborn, David D. Bourgin, Michael Chang, Thomas L. Griffiths
2018 arXiv   pre-print
We introduce a novel problem-solving paradigm that links problem solving and program induction under the Markov Decision Process (MDP) framework.  ...  The importance of hierarchically structured representations for tractable planning has long been acknowledged.  ...  Background The theory of Markov decision processes (MDPs; Sutton & Barto, 1998 ) provides a computational framework for modeling problem solving as search within a metaphorical problem or state space.  ... 
arXiv:1807.07134v1 fatcat:mrubjdh5gveubbj4cblx25c5tm

A modularized framework for explaining hierarchical attention networks on text classifiers

Mahtab Sarvmaili, Amilcar Soares, Riccardo Guidotti, Anna Monreale, Fosca Giannotti, Dino Pedreschi, Stan Matwin
2021 Proceedings of the Canadian Conference on Artificial Intelligence  
In this paper, we propose FEHAN, a modularized Framework for Explaining HiErarchical Attention Network trained to classify text data.  ...  The last decade has witnessed the rise of a black box society where classification models that hide the logic of their internal decision processes are widely adopted due to their high accuracy.  ...  Acknowledgements The authors would like to thank NSERC (Natural Sciences and Engineering Research Council of Canada) for financial support.  ... 
doi:10.21428/594757db.23db72bf fatcat:w2vtraapmzc43m53x755572ehi

Subgoal Identifications in Reinforcement Learning: A Survey [chapter]

Chung-Cheng Chiu, Von-Wun Soo
2011 Advances in Reinforcement Learning  
In sequential decision making, a common approach is to model the entire task with Markov decision processes (MDPs).  ...  The semi-Markov decision process (SMDP) is a framework that extend MDPs to consider temporal-effect.  ... 
doi:10.5772/13214 fatcat:wxraqw43xng6nnxe5qu5diukz4

Modeling Physicians' Utterances to Explore Diagnostic Decision-making

Xuan Guo, Rui Li, Qi Yu, Anne Haake
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
We propose a hierarchical probabilistic framework to learn domain-specific patterns from the medical concepts in these narratives.  ...  These meaningful patterns uncover physicians' diagnostic decision-making processes while parsing the image content.  ...  We would like to thank the participating physicians, the reviewers, and Logical Images, Inc. for images.  ... 
doi:10.24963/ijcai.2017/517 dblp:conf/ijcai/GuoLYH17 fatcat:cmdvnfrfbvds5ldy4asimdzq44
« Previous Showing results 1 — 15 out of 21,007 results