Filters








16,562 Hits in 5.6 sec

Learning goal hierarchies from structured observations and expert annotations

Tolga Könik, John E. Laird
2006 Machine Learning  
Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background  ...  Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to  ...  This allows the use of structured behavior observations represented as temporally changing relational structures, parametric and hierarchical goal annotations, and complex background knowledge.  ... 
doi:10.1007/s10994-006-7734-8 fatcat:birp7qxhtfcuvc2olktcsdkzxu

Learning Goal Hierarchies from Structured Observations and Expert Annotations [chapter]

Tolga Könik, John Laird
2004 Lecture Notes in Computer Science  
Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background  ...  Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to  ...  This allows the use of structured behavior observations represented as temporally changing relational structures, parametric and hierarchical goal annotations, and complex background knowledge.  ... 
doi:10.1007/978-3-540-30109-7_17 fatcat:ctgsu5tbanchboj74qsxs74aky

Learning procedural knowledge through observation

Michael van Lent, John E. Laird
2001 Proceedings of the international conference on Knowledge capture - K-CAP 2001  
The research presented here describes a framework that provides the necessary infrastructure to learn procedural knowledge from observation traces annotated with goal transition information.  ...  One instance of a learning-by-observation system, called KnoMic (Knowledge Mimic), is developed within this framework and evaluated in a complex domain.  ...  These goals correspond to the operator goals in the learned knowledge. The expert can work out a goal hierarchy for the task in advance to aid in specifying goal transitions.  ... 
doi:10.1145/500742.500765 fatcat:4mwozj53kzhwrjejj5im43t5qi

Learning procedural knowledge through observation

Michael van Lent, John E. Laird
2001 Proceedings of the international conference on Knowledge capture - K-CAP 2001  
The research presented here describes a framework that provides the necessary infrastructure to learn procedural knowledge from observation traces annotated with goal transition information.  ...  One instance of a learning-by-observation system, called KnoMic (Knowledge Mimic), is developed within this framework and evaluated in a complex domain.  ...  These goals correspond to the operator goals in the learned knowledge. The expert can work out a goal hierarchy for the task in advance to aid in specifying goal transitions.  ... 
doi:10.1145/500737.500765 dblp:conf/kcap/LentL01 fatcat:giebbiknarc3zmhvn3u2wtvsnm

MineRL: A Large-Scale Dataset of Minecraft Demonstrations [article]

William H. Guss, Brandon Houghton, Nicholay Topin, Phillip Wang, Cayden Codel, Manuela Veloso, Ruslan Salakhutdinov
2019 arXiv   pre-print
However, existing datasets compatible with reinforcement learning simulators do not have sufficient scale, structure, and quality to enable the further development and evaluation of methods focused on  ...  The sample inefficiency of standard deep reinforcement learning methods precludes their application to many real-world problems.  ...  conversations and support.  ... 
arXiv:1907.13440v1 fatcat:63khufur7nd73fb5b43f5jf5gm

Modeling Framework Used to Analyze and Describe Junctional Tourniquet Skills

Benjamin R Bauchwitz, Taylor Curley, Calvin Kwan, James M Niehaus, Carla M Pugh, Peter W Weyhrauch
2019 Military medicine  
We recorded 46 medical first responders performing training exercises applying a junctional tourniquet and used coded video and sensor data to identify the hierarchy of actions they performed in the process  ...  This methodology can improve medical training simulations by indicating changes during the course of learning a new task, such as helpful deviations from instructional protocol.  ...  Parts of the tree corresponding to the high-level goal hierarchy were annotated with Assessment Metrics, Learning Curves, Skills, Knowledge Requirements, and Instructional Material.  ... 
doi:10.1093/milmed/usy348 fatcat:4ypsv33gxvdjrdvl6bxcrtxjyq

Planning with Abstract Learned Models While Learning Transferable Subtasks

John Winder, Stephanie Milani, Matthew Landen, Erebus Oh, Shane Parr, Shawn Squire, Marie DesJardins, Cynthia Matuszek
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
By representing subtasks symbolically using a new formal structure, the lifted abstract Markov decision process (L-AMDP), PALM learns models that are independent and modular.  ...  Through our experiments, we show how PALM integrates planning and execution, facilitating a rapid and efficient learning of abstract, hierarchical models.  ...  IIS-1426452 and by DARPA under grants W911NF-15-1-0503 and D15AP00102.  ... 
doi:10.1609/aaai.v34i06.6555 fatcat:tclwwooz6repdfouimafrgbxwy

Planning with Abstract Learned Models While Learning Transferable Subtasks [article]

John Winder, Stephanie Milani, Matthew Landen, Erebus Oh, Shane Parr, Shawn Squire, Marie desJardins, Cynthia Matuszek
2020 arXiv   pre-print
By representing subtasks symbolically using a new formal structure, the lifted abstract Markov decision process (L-AMDP), PALM learns models that are independent and modular.  ...  Through our experiments, we show how PALM integrates planning and execution, facilitating a rapid and efficient learning of abstract, hierarchical models.  ...  IIS-1813223 and Grant No. IIS-1426452, and by DARPA under grants W911NF-15-1-0503 and D15AP00102.  ... 
arXiv:1912.07544v2 fatcat:m3fyipxdnnc7ppwmsio7wui22e

Learning hierarchical task network domains from partially observed plan traces

Hankz Hankui Zhuo, Héctor Muñoz-Avila, Qiang Yang
2014 Artificial Intelligence  
HTNLearn can learn methods and action models simultaneously from partially observed plan traces (i.e., plan traces where the intermediate states are partially observable).  ...  Encoding such knowledge is a difficult and time-consuming process, even for domain experts.  ...  Qiang Yang is also supported by Hong Kong RGC grants 621011 and 620812.  ... 
doi:10.1016/j.artint.2014.04.003 fatcat:nvuecit6izb2hp4vrg5p6wn6lm

Behavior Bounding: An Efficient Method for High-Level Behavior Comparison

S. A. Wallace
2009 The Journal of Artificial Intelligence Research  
We argue that relatively low amounts of human effort are required to build, maintain, and use the data structures that underlie behavior bounding, and we provide a theoretical basis for these arguments  ...  We show that behavior bounding can be used to compactly represent both human and agent behavior.  ...  Learning By Observation A number of systems (e.g, van Lent & Laird, 1999; Wang, 1995; Konik & Laird, 2006) have also been developed to learn procedural rules or plan operators from observations of expert  ... 
doi:10.1613/jair.2646 fatcat:ekd5u6re2vd45dd6vw2kgmqmcu

Semi-automatic task recognition for interactive narratives with EAT & RUN

Jeff Orkin, Tynan Smith, Hilke Reckman, Deb Roy
2010 Proceedings of the Intelligent Narrative Technologies III Workshop on - INT3 '10  
We describe a semiautomatic methodology for recognizing tasks in gameplay traces, including an annotation tool for non-experts, and a runtime algorithm.  ...  Mining data from online games provides a potential alternative to programming behavior and dialogue for characters in interactive narratives by hand.  ...  structure from the annotations.  ... 
doi:10.1145/1822309.1822312 dblp:conf/int-ws/OrkinSRR10 fatcat:tpf7kj5e4fgj3dwxeoqesc4z34

The 2D shape structure dataset: A user annotated open access database

Axel Carlier, Kathryn Leonard, Stefanie Hahmann, Geraldine Morin, Misha Collins
2016 Computers & graphics  
This database reinforces a philosophy that understanding shape structure as a whole, rather than in the separated categories of parts decomposition, parts hierarchy, and analysis of relationships between  ...  In this paper we present the 2D Shape Structure database, a public, user-generated dataset of 2D shape decompositions into a hierarchy of shape parts with geometric relationships retained.  ...  Acknowledgements: The authors gratefully acknowledge the support of Marie-Paule Cani and advanced grant no. 291184 EXPRESSIVE from the ERC (ERC-2011-ADG 20110209), support from NSF award IIS-0954256, and  ... 
doi:10.1016/j.cag.2016.05.009 fatcat:6q6blnrhivgzvoxxkulxrn3n5y

Multi-Level Discovery of Deep Options [article]

Roy Fox, Sanjay Krishnan, Ion Stoica, Ken Goldberg
2017 arXiv   pre-print
We also show that DDO can discover structure in robot-assisted surgical videos and kinematics that match expert annotation with 72% accuracy.  ...  The scalability of our approach to multi-level hierarchies stems from the decoupling of low-level option discovery from high-level meta-control policy learning, facilitated by under-parametrization of  ...  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Sponsors.  ... 
arXiv:1703.08294v2 fatcat:voytyqqx5vhiliflgay5vge3xi

Fine-Grained Entity Typing in Hyperbolic Space

Federico López, Benjamin Heinzerling, Michael Strube
2019 Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP-2019)  
We evaluate on two datasets and investigate two different techniques for creating a large hierarchical entity type inventory: from an expert-generated ontology and by automatically mining type co-occurrences  ...  We study the ability of hyperbolic embeddings to capture hierarchical relations between mentions in context and their target types in a shared vector space.  ...  Acknowledgments We would like to thank the anonymous reviewers for their valuable comments and suggestions, and we also thank Ana Marasović, Mareike Pfeil, Todor Mihaylov and Mark-Christoph Müller for  ... 
doi:10.18653/v1/w19-4319 dblp:conf/rep4nlp/LopezHS19 fatcat:6qwsmsqsg5galbm3dhyvcetru4

CcNav: Understanding Compiler Optimizations in Binary Code [article]

Sabin Devkota, Pascal Aschwanden, Adam Kunen, Matthew Legendre, Katherine E. Isaacs
2020 arXiv   pre-print
We evaluate CcNav through guided sessions and semi-structured interviews.  ...  Through interviews, feedback, and pair-analytics sessions, we analyzed their tasks and workflow.  ...  As they were able to correlate with source, they annotated the disassembly with variables and structures from source along with evidence of optimizations.  ... 
arXiv:2009.00956v1 fatcat:gi7qv34kfba4hagxq5hr2xc6oe
« Previous Showing results 1 — 15 out of 16,562 results