Partitioning Sensorimotor Space by Predictability Principle in Intrinsic Motivation Systems

Sener, Melisa Idil; Uğur, Emre

doi:10.1109/devlrn.2018.8760504

Cited by 4 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Selection between the regions was made by considering the learning progress IM signal. In our previous work [37], similar to [14], we partitioned the sensorimotor space by considering a single parameter at each partitioning step in order to form exploration regions. Forestier et al [38] developed an algorithmic procedure called "intrinsically motivated goal exploration processes" (IMGEP) that allow the autonomous discovery of skills in an openended learning setting.…”

Section: A Computational Models Of Intrinsic Motivationmentioning

confidence: 99%

Exploration with Intrinsic Motivation using Object-Action-Outcome Latent Space

Sener¹,

Nagai²,

Öztop³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

One effective approach for equipping artificial agents with sensorimotor skills is to use self-exploration. To do this efficiently is critical as time and data collection are costly. In this study, we propose an exploration mechanism that blends action, object, and action outcome representations into a latent space, where local regions are formed to host forward model learning. The agent uses intrinsic motivation to select the forward model with the highest learning progress to adapt at a given exploration step. This parallels how infants learn, as high learning progress indicates that the learning problem is neither too easy nor too difficult in the selected region. The proposed approach is validated with a simulated robot in a table-top environment. The robot interacts with different kinds of objects using a set of parameterized actions and learns the outcomes of these interactions. With the proposed approach, the robot organizes its own curriculum of learning as in existing intrinsic motivation approaches and outperforms them in terms of learning speed. Moreover, the learning regime demonstrates features that partially match infant development, in particular, the proposed system learns to predict grasp action outcomes earlier than that of push action.

show abstract

Section: A Computational Models Of Intrinsic Motivationmentioning

confidence: 99%