Filters








58,890 Hits in 7.3 sec

From Skills to Symbols: Learning Symbolic Representations for Abstract High-Level Planning

George Konidaris, Leslie Pack Kaelbling, Tomas Lozano-Perez
2018 The Journal of Artificial Intelligence Research  
, and a practical mechanism for autonomously learning abstract high-level representations.  ...  We define the specific collection of sets that is necessary and sufficient for planning, and use them to construct a grounded abstract symbolic representation that is provably suitable for deterministic  ...  Acknowledgments The authors would like to thank Alejandro Perez, Patrick Barragán, and Lawson Wong for their critical assistance in designing the robot experiment, and Cameron Allen, Barrett Ames, Garrett  ... 
doi:10.1613/jair.5575 fatcat:sjchgaf6m5a4tn7icvz7p6p3q4

Robots, skills, and symbols

George Konidaris
2013 Proceedings of the 2nd Workshop on Machine Learning for Interactive Systems Bridging the Gap Between Perception, Action and Communication - MLIS '13  
; and on the symbolic representation of plans composed of sequences of skills.  ...  This extended abstract summarizes recent work on skill acquisition, which shows that autonomous robot skill acquisition is feasible, and that a robot can thereby improve its own problem-solving capabilities  ...  This allows an agent to learn a symbolic representation of the world that is suitable for planning using its skills.  ... 
doi:10.1145/2493525.2493528 dblp:conf/ijcai/Konidaris13 fatcat:iqgtacoiifafveq5iaxkoxrml4

Learning Portable Symbolic Representations

Steven James
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
We intend to use these symbolic models to build even higher levels of abstraction, creating a hierarchical representation which could be used to solve complex tasks.  ...  To date, however, research has investigated learning such representations for a single specific task. Our research focuses on approaches to learning these models in a domain-independent manner.  ...  If we have access to skills in this new abstraction, we can then learn symbols once more, creating higher and higher levels of abstraction.  ... 
doi:10.24963/ijcai.2018/826 dblp:conf/ijcai/James18 fatcat:7bme73ud4bhpna5g43dzuqm6q4

From implicit skills to explicit knowledge: a bottom-up model of skill learning

R SUN, E MERRILL, T PETERSON
2001 Cognitive Science  
Different from existing models of mostly high-level skill learning that use a top-down approach (that is, turning declarative knowledge into procedural knowledge through practice), we adopt a bottom-up  ...  Our model is formed by integrating connectionist, reinforcement, and symbolic learning methods to perform on-line reactive learning.  ...  We wish to thank Susan Chipman and Helen Gigley for the support, Jeff Shrager, Jack Gelfand, Dave Waltz, and Diana Gordon for comments or discussions, and Jim Ballas and Alan Schultz for assistance with  ... 
doi:10.1016/s0364-0213(01)00035-0 fatcat:v77o32qy7vd5fko5drlgwfez74

From implicit skills to explicit knowledge: a bottom-up model of skill learning

Ron Sun, Edward Merrill, Todd Peterson
2001 Cognitive Science  
Different from existing models of mostly high-level skill learning that use a top-down approach (that is, turning declarative knowledge into procedural knowledge through practice), we adopt a bottom-up  ...  Our model is formed by integrating connectionist, reinforcement, and symbolic learning methods to perform on-line reactive learning.  ...  We wish to thank Susan Chipman and Helen Gigley for the support, Jeff Shrager, Jack Gelfand, Dave Waltz, and Diana Gordon for comments or discussions, and Jim Ballas and Alan Schultz for assistance with  ... 
doi:10.1207/s15516709cog2502_2 fatcat:32nltgsmmrhjtpze6vpkpuqvrq

Learning High-Level Planning Symbols from Intrinsically Motivated Experience

Angelo Oddi, Riccardo Rasconi, Emilio Cartoni, Gabriele Sartor, Gianluca Baldassarre, Vieri Giuliano Santucci
2019 Zenodo  
For instance, in [3] an algorithm is presented for automatically producing symbolic domains based on the Planning Domain Definition Language (PDDL, see [2]), starting from a set of low-level skills represented  ...  Such information abstraction process indeed reveals invaluable for high-level planning, as it allows to make explicit the causal relations existing at the high-level which would otherwise remain hidden  ...  Such information abstraction process indeed reveals invaluable for high-level planning, as it allows to make explicit the causal relations existing at the high-level which would otherwise remain hidden  ... 
doi:10.5281/zenodo.4785261 fatcat:lz6u7jz24rhbll3wmbo6ul72mu

Constructing Abstraction Hierarchies Using a Skill-Symbol Loop [article]

George Konidaris
2015 arXiv   pre-print
We describe a framework for building abstraction hierarchies whereby an agent alternates skill- and representation-acquisition phases to construct a sequence of increasingly abstract Markov decision processes  ...  Our formulation builds on recent results showing that the appropriate abstract representation of a problem is specified by the agent's skills.  ...  We show that these two processes can be combined into a skill-symbol loop: the agent acquires a set of high-level skills, then constructs the appropriate representation for planning using them, resulting  ... 
arXiv:1509.07582v1 fatcat:he6cxbuyqrfbbn2md7yf6zhusi

Creative Symbolic Interaction

Gérard Assayag
2014 Proceedings of the SMC Conferences  
(Abstract to follow)  ...  Acknowledgments Thanks to the other OMax Brothers: M. Chemillier, S. Dubnov, G. Bloch, B. Lévy, L. Bonnasse-Gahot, J. Nika.  ...  Learning Skills Interactive learning of musical structures stems from data provided by the sequential process of listening, learning symbolic models that capture high-level multi-dimensional and multi-scale  ... 
doi:10.5281/zenodo.850441 fatcat:arpsr4vnd5ddhmha4vcyw4e2v4

Constructing Abstraction Hierarchies Using a Skill-Symbol Loop

George Konidaris
2016 IJCAI International Joint Conference on Artificial Intelligence  
We describe a framework for building abstraction hierarchies whereby an agent alternates skill- and representation-construction phases to construct a sequence of increasingly abstract Markov decision processes  ...  Our formulation builds on recent results showing that the appropriate abstract representation of a problem is specified by the agent's skills.  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon.  ... 
pmid:28579718 pmcid:PMC5455777 fatcat:3lmyu7qmbbgotkzxnw2wpwpe6q

Active Exploration for Learning Symbolic Representations [article]

Garrett Andersen, George Konidaris
2017 arXiv   pre-print
We introduce an online active exploration algorithm for data-efficiently learning an abstract symbolic model of an environment.  ...  Our algorithm is divided into two parts: the first part quickly generates an intermediate Bayesian symbolic model from the data that the agent has collected so far, which the agent can then use along with  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon.  ... 
arXiv:1709.01490v2 fatcat:2ehvqcf35vg7fnm4r7y33xkqjm

Tangible symbols: symbolic communication for individuals with multisensory impairments

Charity Rowland, Philip Schweigert
1989 Augmentative and Alternative Communication : AAC  
Chapters corer such issues as how to construct tangible symbol systems, how to implement their use in functional routines, and how to plan short-and long-term communication programs that will ensure progress  ...  This manual is intended to accompany a videotape entitled "Tangible Symbol Systems."  ...  Once an individual is ready to learn a new type of symbol, move to a level of representation that is more abstract, more conventional, and/or more portable.  ... 
doi:10.1080/07434618912331275276 fatcat:xvwg2nz2znbd7j44nzbmgtrwkq

Mapping and predicting literacy and reasoning skills from early to later primary school

Andreas Demetriou, Christine Merrell, Peter Tymms
2017 Learning and Individual Differences  
2017) 'Mapping and predicting literacy and reasoning skills from early to later primary school.', Learning and individual dierences., 54 . pp. 217-225.  ...  The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes  ...  In the first phase of this cycle, from 2 to 4 years, actionbased episodic representations of the previous cycle are elevated into symbol-based mental representations.  ... 
doi:10.1016/j.lindif.2017.01.023 fatcat:q7yokn6xnnbaxgg4ifijmh3yvm

Leveraging Approximate Symbolic Models for Reinforcement Learning via Skill Diversity [article]

Lin Guan, Sarath Sreedharan, Subbarao Kambhampati
2022 arXiv   pre-print
We will use these models to extract high-level landmarks that will be used to decompose the task, and at the low level, we learn a set of diverse policies for each possible task sub-goal identified by  ...  While previous works have looked at the possibility of using symbolic models along with RL approaches, they tend to assume that the high-level action models are executable at low level and the fluents  ...  We can't identify this from the high-level symbolic information alone, since it contains no information about the battery level.  ... 
arXiv:2202.02886v2 fatcat:s44xca22pvevnbuzgyx4xh2rse

Transferring skills to humanoid robots by extracting semantic representations from observations of human activities

Karinne Ramirez-Amaro, Michael Beetz, Gordon Cheng
2017 Artificial Intelligence  
A pioneering study of high level representations was introduced by Kuniyoshi et al.  ...  [35] , employed a (partially) symbolic representation of manipulation strategies to generate robot plans, based on preand post-conditions.  ...  Supplementary material Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.artint.2015.08.009.  ... 
doi:10.1016/j.artint.2015.08.009 fatcat:or6jvri7jnfurhbh2odltkmx4a

Perceptual symbol systems

L W Barsalou
1999 Behavioral and Brain Sciences  
The storage and reactivation of perceptual symbols operates at the level of perceptual components--not at the level of holistic perceptual experiences.  ...  Propositions result from binding simulators to perceived individuals to represent type-token relations.  ...  I am also grateful to Howard Nusbaum for discussion on cognitive penetration, and to Gay Snodgrass for permission to use drawings from the Snodgrass and Vanderwart (1980) norms.  ... 
pmid:11301525 fatcat:z6plr3yebfcfhgv664m6mg53zy
« Previous Showing results 1 — 15 out of 58,890 results