An architecture for observational learning and decision making based on internal models

Antonio Chella, Haris Dindo, Giuseppe La Tona, Kristinn R. Thórisson, Eric Nivel

Risultato della ricerca: Article

1 Citazione (Scopus)

Abstract

We present a cognitive architecture whose main constituents are allowed to grow through a situated experience in the world. Such an architectural growth is bootstrapped from a minimal initial knowledge and the architecture itself is built around the biologically-inspired notion of internal models. The key idea, supported by findings in cognitive neuroscience, is that the same internal models used in overt goal-directed action execution can be covertly re-enacted in simulation to provide a unifying explanation to a number of apparently unrelated individual and social phenomena, such as state estimation, action and intention understanding, imitation learning and mindreading. Thus, rather than reasoning over abstract symbols, we rely on the biologically plausible processes firmly grounded in the actual sensorimotor experience of the agent. The article describes how such internal models are learned in the first place, either through individual experience or by observing and imitating other skilled agents, and how they are used in action planning and execution. Furthermore, we explain how the architecture continuously adapts its internal agency and how increasingly complex cognitive phenomena, such as continuous learning, prediction and anticipation, result from an interplay of simpler principles. We describe an early evaluation of our approach in a classical AI problem-solving domain: the Sokoban puzzle.
Lingua originaleEnglish
pagine (da-a)52-63
Numero di pagine12
RivistaBiologically Inspired Cognitive Architectures
Volume5
Stato di pubblicazionePublished - 2013

Fingerprint

Decision Making
Decision making
Learning
State estimation
Growth
Planning
Cognitive Neuroscience

All Science Journal Classification (ASJC) codes

  • Experimental and Cognitive Psychology
  • Artificial Intelligence
  • Cognitive Neuroscience

Cita questo

@article{d5ad36a54b274eeeb05ea0deeefe7454,
title = "An architecture for observational learning and decision making based on internal models",
abstract = "We present a cognitive architecture whose main constituents are allowed to grow through a situated experience in the world. Such an architectural growth is bootstrapped from a minimal initial knowledge and the architecture itself is built around the biologically-inspired notion of internal models. The key idea, supported by findings in cognitive neuroscience, is that the same internal models used in overt goal-directed action execution can be covertly re-enacted in simulation to provide a unifying explanation to a number of apparently unrelated individual and social phenomena, such as state estimation, action and intention understanding, imitation learning and mindreading. Thus, rather than reasoning over abstract symbols, we rely on the biologically plausible processes firmly grounded in the actual sensorimotor experience of the agent. The article describes how such internal models are learned in the first place, either through individual experience or by observing and imitating other skilled agents, and how they are used in action planning and execution. Furthermore, we explain how the architecture continuously adapts its internal agency and how increasingly complex cognitive phenomena, such as continuous learning, prediction and anticipation, result from an interplay of simpler principles. We describe an early evaluation of our approach in a classical AI problem-solving domain: the Sokoban puzzle.",
author = "Antonio Chella and Haris Dindo and {La Tona}, Giuseppe and Th{\'o}risson, {Kristinn R.} and Eric Nivel",
year = "2013",
language = "English",
volume = "5",
pages = "52--63",
journal = "Biologically Inspired Cognitive Architectures",
issn = "2212-683X",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - An architecture for observational learning and decision making based on internal models

AU - Chella, Antonio

AU - Dindo, Haris

AU - La Tona, Giuseppe

AU - Thórisson, Kristinn R.

AU - Nivel, Eric

PY - 2013

Y1 - 2013

N2 - We present a cognitive architecture whose main constituents are allowed to grow through a situated experience in the world. Such an architectural growth is bootstrapped from a minimal initial knowledge and the architecture itself is built around the biologically-inspired notion of internal models. The key idea, supported by findings in cognitive neuroscience, is that the same internal models used in overt goal-directed action execution can be covertly re-enacted in simulation to provide a unifying explanation to a number of apparently unrelated individual and social phenomena, such as state estimation, action and intention understanding, imitation learning and mindreading. Thus, rather than reasoning over abstract symbols, we rely on the biologically plausible processes firmly grounded in the actual sensorimotor experience of the agent. The article describes how such internal models are learned in the first place, either through individual experience or by observing and imitating other skilled agents, and how they are used in action planning and execution. Furthermore, we explain how the architecture continuously adapts its internal agency and how increasingly complex cognitive phenomena, such as continuous learning, prediction and anticipation, result from an interplay of simpler principles. We describe an early evaluation of our approach in a classical AI problem-solving domain: the Sokoban puzzle.

AB - We present a cognitive architecture whose main constituents are allowed to grow through a situated experience in the world. Such an architectural growth is bootstrapped from a minimal initial knowledge and the architecture itself is built around the biologically-inspired notion of internal models. The key idea, supported by findings in cognitive neuroscience, is that the same internal models used in overt goal-directed action execution can be covertly re-enacted in simulation to provide a unifying explanation to a number of apparently unrelated individual and social phenomena, such as state estimation, action and intention understanding, imitation learning and mindreading. Thus, rather than reasoning over abstract symbols, we rely on the biologically plausible processes firmly grounded in the actual sensorimotor experience of the agent. The article describes how such internal models are learned in the first place, either through individual experience or by observing and imitating other skilled agents, and how they are used in action planning and execution. Furthermore, we explain how the architecture continuously adapts its internal agency and how increasingly complex cognitive phenomena, such as continuous learning, prediction and anticipation, result from an interplay of simpler principles. We describe an early evaluation of our approach in a classical AI problem-solving domain: the Sokoban puzzle.

UR - http://hdl.handle.net/10447/90189

UR - http://linkinghub.elsevier.com/retrieve/pii/S2212683X1300042X

M3 - Article

VL - 5

SP - 52

EP - 63

JO - Biologically Inspired Cognitive Architectures

JF - Biologically Inspired Cognitive Architectures

SN - 2212-683X

ER -