• Sorted by Date • Classified by Publication Type • Classified by Topic • Sorted by First Author Last Name •
Ori Ossmy, Danyang Han, Patrick MacAlpine, Justine Hoch, Peter Stone, and Karen E. Adolph. Walking and falling: Using robot simulations to model the role of errors in infant walking. Developmental Science, 27:e13449, September 2023.
Available from the publisher's webpage
What is the optimal penalty for errors in infant skill learning? Behavioral analyses indicate that errors are frequent but trivial as infants acquire foundational skills. In learning to walk, for example, falling is commonplace but appears to incur only a negligible penalty. Behavioral data, however, cannot reveal whether a low penalty for falling is beneficial for learning to walk. Here, we used a simulated bipedal robot as an embodied model to test the optimal penalty for errors in learning to walk. We trained the robot to walk using 12,500 independent simulations on walking paths produced by infants during free play and systematically varied the penalty for falling -- a level of precision, control, and magnitude impossible with real infants. When trained with lower penalties for falling, the robot learned to walk farther and better on familiar, trained paths and better generalized its learning to novel, untrained paths. Indeed, zero penalty for errors led to the best performance for both learning and generalization. Moreover, the beneficial effects of a low penalty were stronger for generalization than for learning. Robot simulations corroborate prior behavioral data and suggest that a low penalty for errors helps infants learn foundational skills (e.g., walking, talking, and social interactions) that require immense flexibility, creativity, and adaptability. Research Highlights During infant skill acquisition, errors are commonplace but appear to incur a low penalty; when learning to walk, for example, falls are frequent but trivial. To test the optimal penalty for errors, we trained a simulated robot to walk using real infant paths and systematically manipulated the penalty for falling. Lower penalties in training led to better performance on familiar, trained paths and on novel untrained paths, and zero penalty was most beneficial. Benefits of a low penalty were stronger for untrained than for trained paths, suggesting that discounting errors facilitates acquiring skills that require immense flexibility and generalization.
@article{Ossmy2023,
author = {Ossmy, Ori and Han, Danyang and MacAlpine, Patrick and Hoch, Justine and Stone, Peter and Adolph, Karen E.},
title = {Walking and falling: Using robot simulations to model the role of errors in infant walking},
journal = {Developmental Science},
pages = {e13449},
volume={27},
issue={2},
year="2023",
month="September",
keywords = {error, falling, penalty, reinforcement learning, simulated robot, walking},
doi = {https://doi.org/10.1111/desc.13449},
url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/desc.13449},
eprint = {https://onlinelibrary.wiley.com/doi/pdf/10.1111/desc.13449},
abstract = {
What is the optimal penalty for errors in infant skill
learning? Behavioral analyses indicate that errors are
frequent but trivial as infants acquire foundational
skills. In learning to walk, for example, falling is
commonplace but appears to incur only a negligible
penalty. Behavioral data, however, cannot reveal whether
a low penalty for falling is beneficial for learning to
walk. Here, we used a simulated bipedal robot as an
embodied model to test the optimal penalty for errors in
learning to walk. We trained the robot to walk using
12,500 independent simulations on walking paths produced
by infants during free play and systematically varied
the penalty for falling -- a level of precision, control,
and magnitude impossible with real infants. When trained
with lower penalties for falling, the robot learned to
walk farther and better on familiar, trained paths and
better generalized its learning to novel, untrained
paths. Indeed, zero penalty for errors led to the best
performance for both learning and
generalization. Moreover, the beneficial effects of a
low penalty were stronger for generalization than for
learning. Robot simulations corroborate prior behavioral
data and suggest that a low penalty for errors helps
infants learn foundational skills (e.g., walking,
talking, and social interactions) that require immense
flexibility, creativity, and adaptability. Research
Highlights During infant skill acquisition, errors are
commonplace but appear to incur a low penalty; when
learning to walk, for example, falls are frequent but
trivial. To test the optimal penalty for errors, we
trained a simulated robot to walk using real infant
paths and systematically manipulated the penalty for
falling. Lower penalties in training led to better
performance on familiar, trained paths and on novel
untrained paths, and zero penalty was most
beneficial. Benefits of a low penalty were stronger for
untrained than for trained paths, suggesting that
discounting errors facilitates acquiring skills that
require immense flexibility and generalization.},
wwwnote = {Available from the <a href="https://onlinelibrary.wiley.com/doi/abs/10.1111/desc.13449">publisher's webpage</a>},
}
Generated by bib2html.pl (written by Patrick Riley ) on Tue Nov 19, 2024 10:29:29