Corpus ID: 47017748

Deep Curiosity Loops in Social Environments

@article{Barkan2018DeepCL,
  title={Deep Curiosity Loops in Social Environments},
  author={Jonatan Barkan and Goren Gordon},
  journal={ArXiv},
  year={2018},
  volume={abs/1806.03645}
}
  • Jonatan Barkan, Goren Gordon
  • Published in ArXiv 2018
  • Computer Science
  • Inspired by infants' intrinsic motivation to learn, which values informative sensory channels contingent on their immediate social environment, we developed a deep curiosity loop (DCL) architecture. The DCL is composed of a learner, which attempts to learn a forward model of the agent's state-action transition, and a novel reinforcement-learning (RL) component, namely, an Action-Convolution Deep Q-Network, which uses the learner's prediction error as reward. The environment for our agent is… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    3
    Twitter Mentions

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 43 REFERENCES

    Intrinsic Motivation Systems for Autonomous Mental Development

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Efficient human face detection in infancy.

    VIEW 1 EXCERPT

    Face Detection with the Faster R-CNN

    VIEW 2 EXCERPTS

    Nesterov's accelerated gradient and momentum as approximations to regularised update descent

    VIEW 1 EXCERPT

    Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

    VIEW 1 EXCERPT