Filters








988 Hits in 6.2 sec

Identifying Reasoning Flaws in Planning-Based RL Using Tree Explanations [article]

Kin-Ho Lam, Zhengxian Lin, Jed Irvine, Jonathan Dodge, Zeyad T Shureih, Roli Khanna, Minsuk Kahng, Alan Fern
2021 arXiv   pre-print
We consider identifying such flaws in a planning-based deep reinforcement learning (RL) agent for a complex real-time strategy game.  ...  This gives the potential for humans to identify flaws at the level of reasoning steps in the tree, even if the entire reasoning process is too complex to understand.  ...  Planning-Based RL Agent We describe our planning-based RL architecture and learning approach, where decisions are made via tree search using a learned game model and evaluation function.  ... 
arXiv:2109.13978v1 fatcat:srck72u6orc4tk7pgz75xlbyam

Explainable Goal-Driven Agents and Robots – A Comprehensive Review [article]

Fatai Sado, Chu Kiong Loo, Wei Shiung Liew, Matthias Kerzel, Stefan Wermter
2021 arXiv   pre-print
(example, beliefs, desires, intention, plans, and goals) with humans in the loop.  ...  reviews approaches on explainable goal-driven intelligent agents and robots, focusing on techniques for explaining and communicating agents perceptual functions (example, senses, and vision) and cognitive reasoning  ...  the agent's cognitive base, the cognitive base (dominated by deliberative reasoning) relates plans, goals, beliefs, or desires to executed actions.  ... 
arXiv:2004.09705v7 fatcat:p5jxv5hfk5elphzre4cn6acgsa

Explainable artificial intelligence for autonomous driving: An overview and guide for future research directions [article]

Shahin Atakishiyev, Mohammad Salameh, Hengshuai Yao, Randy Goebel
2022 arXiv   pre-print
However, intelligent decision-making in autonomous cars is not generally understandable by humans in the current state of the art, and such deficiency hinders this technology from being socially acceptable  ...  Autonomous driving has achieved a significant milestone in research and development over the last decade.  ...  ], 2021 Enhancing automated driving with human Gaze-based vehicle Visual Road users foresight reference Omeiza et al., [122], 2021 Generating tree-based explanations with and Tree-based represen- Textual  ... 
arXiv:2112.11561v2 fatcat:zluqlvmtznh25eihtouubib3ba

Explainability in reinforcement learning: perspective and position [article]

Agneza Krajna, Mario Brcic, Tomislav Lipic, Juraj Doncevic
2022 arXiv   pre-print
It is therefore essential to improve the trust and transparency of RL-based systems through explanations.  ...  Most articles dealing with explainability in artificial intelligence provide methods that concern supervised learning and there are very few articles dealing with this in the area of RL.  ...  [45] suggest a slightly different method for achieving interpretability of RL policy using trees.  ... 
arXiv:2203.11547v1 fatcat:7zfi7f3i6bdgbhcjg7izvpnieq

A review of learning planning action models

Ankuj Arora, Humbert Fiorino, Damien Pellier, Marc Métivier, Sylvie Pesty
2018 Knowledge engineering review (Print)  
The reasons range from intense laborious effort, to intricacies so barely identifiable, that programming them is a challenge that presents itself much later in the process.  ...  It then details the learning techniques that have been used in the literature during the past decades, and finally presents some open issues.  ...  An explanation component reasons about hidden information in the environment. Finally, a goal management component manages and communicates the goals to the planner.  ... 
doi:10.1017/s0269888918000188 fatcat:yjwlfpz4efhnrbhxczr2mxh4km

Introspective multistrategy learning: On the construction of learning strategies

Michael T. Cox, Ashwin Ram
1999 Artificial Intelligence  
Extensive empirical evaluations of Meta-AQUA show that it performs significantly better in a deliberative, planful mode than in a reflexive mode in which learning goals are ablated and, furthermore, that  ...  of the reasoning failure, the learner posts explicitly represented learning goals to change its background knowledge (deciding what to learn); and given a set of learning goals, the learner uses nonlinear  ...  We also wish to gratefully acknowledge Mark Devaney's assistance in reprogramming Tale-Spin.  ... 
doi:10.1016/s0004-3702(99)00047-8 fatcat:6m2vpnokt5awhmcctnof6csgxi

A review of machine learning for automated planning

Sergio Jiménez, Tomás De La Rosa, Susana Fernández, Fernando Fernández, Daniel Borrajo
2012 Knowledge engineering review (Print)  
This paper reviews recent techniques in machine learning for the automatic definition of planning knowledge.  ...  Recent discoveries in automated planning are broadening the scope of planners, from toy problems to real applications. However, applying automated planners to real-world problems is far from simple.  ...  It simply uses Explanation Based Learning (EBL) to analyze the relations between actions preconditions and effects.  ... 
doi:10.1017/s026988891200001x fatcat:slnkph7hyve3higgkjyd3wydlu

A DPLL(T) Theory Solver for a Theory of Strings and Regular Expressions [chapter]

Tianyi Liang, Andrew Reynolds, Cesare Tinelli, Clark Barrett, Morgan Deters
2014 Lecture Notes in Computer Science  
These techniques can be used to integrate string reasoning into general, multi-theory SMT solvers based on the DPLL(T ) architecture.  ...  We have implemented them in our SMT solver CVC4 to expand its already large set of built-in theories to a theory of strings with concatenation, length, and membership in regular languages.  ...  Acknowledgments We would like to thank Nestan Tsiskaridze for her insightful comments, and the developers of Z3-STR for their technical support in using their tool and several clarifications on it.  ... 
doi:10.1007/978-3-319-08867-9_43 fatcat:35m2qckgs5frbfflsev4uj5rju

Computing Incoherence Explanations for Learned Ontologies [chapter]

Daniel Fleischhacker, Christian Meilicke, Johanna Völker, Mathias Niepert
2013 Lecture Notes in Computer Science  
Off-the-shelf debuggers based on logical reasoning struggle with the particular characteristics of learned ontologies.  ...  They are mostly inefficient when it comes to detecting modeling flaws, or highlighting all of the logical reasons for the discovered problems.  ...  In particular, we are not interested in ABox reasoning, nor do we require many constructs supported by OWL RL.  ... 
doi:10.1007/978-3-642-39666-3_7 fatcat:hzy2a4prjba2pk44cha4laruey

Explanation-based learning: a survey of programs and perspectives

Thomas Ellman
1989 ACM Computing Surveys  
Explanation-based learning (EBL) is a technique by which an intelligent system can learn by observing examples.  ...  This paper provides a general introduction to the field of explanation-based learning. Considerable emphasis is placed on showing how EBL combines the four learning tasks mentioned above.  ...  Modified goal regression (MGR): A procedure used in explanation-based generalization (EBG) to analyze proof trees [Mitchell et al. 19861 .  ... 
doi:10.1145/66443.66445 fatcat:o25qzli5sza3nczyifhhcl2roi

Creative Problem Solving in Artificially Intelligent Agents: A Survey and Framework [article]

Evana Gizzi, Lakshmi Nair, Sonia Chernova, Jivko Sinapov
2022 arXiv   pre-print
post deployment, remains a limiting factor in the safe and useful integration of intelligent systems.  ...  Despite many advancements in planning and learning, resolving novel problems or adapting existing knowledge to a new context, especially in cases where the environment may change in unpredictable ways  ...  In Sarathy et al. (Sarathy et al., 2020) , the agent uses RL in a fully symbolic 2D 'grid-world' domain to resolve failures in PDDL plans. Chitnis et al.  ... 
arXiv:2204.10358v1 fatcat:hiccmw67rjapbo374ucvresaam

Analysis of Student Desertion in a Systems and Computing Engineering Undergraduate Program

Luis Fernando Castro, Esperanza Espitia P., Sergio Augusto Cardona
2019 Revista Colombiana de Computación  
in databases) perspective and using techniques for identifying students' behavioral patterns.  ...  This proposal is important because it will support higher education institutions in decision-making and creating action plans to reduce the high rate of student attrition.  ...  This project is intended to generate a decision tree model by implementing the J48 algorithm, using the WEKA tool to identify such causes.  ... 
doi:10.29375/25392115.3608 fatcat:ifo7tyuxmnbkhhgayktneqt2jy

Learning-Assisted Automated Planning: Looking Back, Taking Stock, Going Forward

Terry Zimmerman, Subbarao Kambhampati
2003 The AI Magazine  
We extend the survey analysis to suggest promising avenues for future research in learning based on both previous experience and current needs in the planning community.  ...  This article reports on an extensive survey and analysis of research work related to machine learning as it applies to automated planning over the past 30 years.  ...  DT = decision tree. EBL = explanation-based learning. IP = integer linear programming. NN = neural network. RL = reinforcement learning. SAT: satisfiability.  ... 
doi:10.1609/aimag.v24i2.1705 dblp:journals/aim/ZimmermanK03 fatcat:5umdmeki4zdlznah2hhdrjpe5e

Reward (Mis)design for Autonomous Driving [article]

W. Bradley Knox, Alessandro Allievi, Holger Banzhaf, Felix Schmitt, Peter Stone
2022 arXiv   pre-print
To diagnose common errors, we develop 8 simple sanity checks for identifying flaws in reward functions.  ...  These sanity checks are applied to reward functions from past work on reinforcement learning (RL) for autonomous driving (AD), revealing near-universal flaws in reward design for AD that might also exist  ...  The terms of this arrangement have been reviewed and approved by the University of Texas at Austin in accordance with its policy on objectivity in research.  ... 
arXiv:2104.13906v2 fatcat:cifcxmfv45efnhpzt4q4ptglt4

A Review on Learning Planning Action Models for Socio-Communicative HRI [article]

Ankuj Arora and Humbert Fiorino and Damien Pellier and Sylvie Pesty
2018 arXiv   pre-print
AI techniques, such as machine learning, can be used to learn behavioral models (also known as symbolic action models in AI), intended to be reusable for AI planning, from the aforementioned multimodal  ...  In the same domain, this action sequence that stems from plan execution can be represented as a trace.  ...  In the literature, AI (or Automated) planning has been used in HRI for robot action planning and reasoning [1] .  ... 
arXiv:1810.09245v1 fatcat:5ga2klfbcfaife7wylepgsxtzq
« Previous Showing results 1 — 15 out of 988 results