Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm

A single shepherd dog can herd a flock of sheep to a gate. Despite a heuristic algorithm of a dog based on adaptive switching between collecting the sheep when they are too dispersed and driving them once they are aggregated, it remains unknown how the dog learns the algorithm of switching. In fact,...

Full description

Saved in:
Bibliographic Details
Main Authors: Go, Clark Kendrick C, Koganti, Nishanth, Ikeda, Kazushi
Format: text
Published: Archīum Ateneo 2021
Subjects:
Online Access:https://archium.ateneo.edu/mathematics-faculty-pubs/180
https://ieeexplore.ieee.org/document/9533722
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Ateneo De Manila University
id ph-ateneo-arc.mathematics-faculty-pubs-1184
record_format eprints
spelling ph-ateneo-arc.mathematics-faculty-pubs-11842022-02-24T00:49:23Z Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm Go, Clark Kendrick C Koganti, Nishanth Ikeda, Kazushi A single shepherd dog can herd a flock of sheep to a gate. Despite a heuristic algorithm of a dog based on adaptive switching between collecting the sheep when they are too dispersed and driving them once they are aggregated, it remains unknown how the dog learns the algorithm of switching. In fact, reinforcement learning models have not succeeded so far in reproducing the switching algorithm without explicitly making two strategies. Here, we show that an imitation learning model can reproduce the switching algorithm, that is, the dog learns the algorithm from demonstrations by an expert. We also confirmed that the dog does not simply copy the demonstrations but learns the required task by showing that it can herd more sheep than those in the given demonstrations. 2021-01-01T08:00:00Z text https://archium.ateneo.edu/mathematics-faculty-pubs/180 https://ieeexplore.ieee.org/document/9533722 Mathematics Faculty Publications Archīum Ateneo Adaptation models Heuristic algorithms Neural networks Switches Dogs Reinforcement learning Logic gates imitation learning herding Mathematics
institution Ateneo De Manila University
building Ateneo De Manila University Library
continent Asia
country Philippines
Philippines
content_provider Ateneo De Manila University Library
collection archium.Ateneo Institutional Repository
topic Adaptation models
Heuristic algorithms
Neural networks
Switches
Dogs
Reinforcement learning
Logic gates
imitation learning
herding
Mathematics
spellingShingle Adaptation models
Heuristic algorithms
Neural networks
Switches
Dogs
Reinforcement learning
Logic gates
imitation learning
herding
Mathematics
Go, Clark Kendrick C
Koganti, Nishanth
Ikeda, Kazushi
Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
description A single shepherd dog can herd a flock of sheep to a gate. Despite a heuristic algorithm of a dog based on adaptive switching between collecting the sheep when they are too dispersed and driving them once they are aggregated, it remains unknown how the dog learns the algorithm of switching. In fact, reinforcement learning models have not succeeded so far in reproducing the switching algorithm without explicitly making two strategies. Here, we show that an imitation learning model can reproduce the switching algorithm, that is, the dog learns the algorithm from demonstrations by an expert. We also confirmed that the dog does not simply copy the demonstrations but learns the required task by showing that it can herd more sheep than those in the given demonstrations.
format text
author Go, Clark Kendrick C
Koganti, Nishanth
Ikeda, Kazushi
author_facet Go, Clark Kendrick C
Koganti, Nishanth
Ikeda, Kazushi
author_sort Go, Clark Kendrick C
title Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
title_short Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
title_full Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
title_fullStr Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
title_full_unstemmed Solving the Shepherding Problem: Imitation Learning Can Acquire the Switching Algorithm
title_sort solving the shepherding problem: imitation learning can acquire the switching algorithm
publisher Archīum Ateneo
publishDate 2021
url https://archium.ateneo.edu/mathematics-faculty-pubs/180
https://ieeexplore.ieee.org/document/9533722
_version_ 1726158632853176320