TY - JOUR
T1 - A Theory of Relation Learning and Cross-Domain Generalization
AU - Doumas, Leonidas A.A.
AU - Puebla, Guillermo
AU - Martin, Andrea E.
AU - Hummel, John E.
N1 - Publisher Copyright:
© 2022 The Author(s)
PY - 2022/2/3
Y1 - 2022/2/3
N2 - People readily generalize knowledge to novel domains and stimuli. We present a theory, instantiated in a computational model, based on the idea that cross-domain generalization in humans is a case of analogical inference over structured (i.e., symbolic) relational representations. The model is an extension of the Learning and Inferencewith Schemas andAnalogy (LISA; Hummel&Holyoak, 1997, 2003) and Discovery of Relations by Analogy (DORA; Doumas et al., 2008) models of relational inference and learning. The resulting model learns both the content and format (i.e., structure) of relational representations from nonrelational inputswithout supervision, when augmented with the capacity for reinforcement learning it leverages these representations to learn about individual domains, and then generalizes to new domains on the first exposure (i.e., zero-shot learning) via analogical inference. We demonstrate the capacity of the model to learn structured relational representations from a variety of simple visual stimuli, and to perform cross-domain generalization between video games (Breakout and Pong) and between several psychological tasks. We demonstrate that the model’s trajectory closely mirrors the trajectory of children as they learn about relations, accounting for phenomena from the literature on the development of children’s reasoning and analogy making. The model’s ability to generalize between domains demonstrates the flexibility afforded by representing domains in terms of their underlying relational structure, rather than simply in terms of the statistical relations between their inputs and outputs.
AB - People readily generalize knowledge to novel domains and stimuli. We present a theory, instantiated in a computational model, based on the idea that cross-domain generalization in humans is a case of analogical inference over structured (i.e., symbolic) relational representations. The model is an extension of the Learning and Inferencewith Schemas andAnalogy (LISA; Hummel&Holyoak, 1997, 2003) and Discovery of Relations by Analogy (DORA; Doumas et al., 2008) models of relational inference and learning. The resulting model learns both the content and format (i.e., structure) of relational representations from nonrelational inputswithout supervision, when augmented with the capacity for reinforcement learning it leverages these representations to learn about individual domains, and then generalizes to new domains on the first exposure (i.e., zero-shot learning) via analogical inference. We demonstrate the capacity of the model to learn structured relational representations from a variety of simple visual stimuli, and to perform cross-domain generalization between video games (Breakout and Pong) and between several psychological tasks. We demonstrate that the model’s trajectory closely mirrors the trajectory of children as they learn about relations, accounting for phenomena from the literature on the development of children’s reasoning and analogy making. The model’s ability to generalize between domains demonstrates the flexibility afforded by representing domains in terms of their underlying relational structure, rather than simply in terms of the statistical relations between their inputs and outputs.
KW - Generalization
KW - Learning relational content
KW - Learning structured representations
KW - Neural oscillations
KW - Relation learning
UR - http://www.scopus.com/inward/record.url?scp=85125066901&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85125066901&partnerID=8YFLogxK
U2 - 10.1037/rev0000346
DO - 10.1037/rev0000346
M3 - Article
C2 - 35113620
AN - SCOPUS:85125066901
SN - 0033-295X
VL - 129
SP - 999
EP - 1041
JO - Psychological review
JF - Psychological review
IS - 5
ER -