Self-Communicating Deep Reinforcement Learning Agents Develop External Number Representations
Peer reviewed, Journal article
Published version
Date
2022Metadata
Show full item recordCollections
- Institutt for lærerutdanning [3729]
- Publikasjoner fra CRIStin - NTNU [38294]
Original version
10.7557/18.6291Abstract
Symbolic numbers are a remarkable product ofhuman cultural development. The developmentalprocess involved the creation and progressive re-finement of material representational tools, suchas notched tallies, knotted strings, and countingboards. In this paper, we introduce a computa-tional framework that allows the investigation ofhow material representations might support num-ber processing in a deep reinforcement learning sce-nario. In this framework, agents can use an exter-nal, discrete state to communicate information tosolve a simple numerical estimation task. We findthat different perceptual and processing constraintsresult in different emergent representations, whosespecific characteristics can facilitate the learningand communication of numbers.