Salvato in:
Dettagli Bibliografici
Autori principali: Chan, Ayshah, Schneider, Maja, Körner, Marco
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2310.06574
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916693343731712
author Chan, Ayshah
Schneider, Maja
Körner, Marco
author_facet Chan, Ayshah
Schneider, Maja
Körner, Marco
contents We propose an approach for early crop classification through identifying important timesteps with eXplainable AI (XAI) methods. Our approach consists of training a baseline crop classification model to carry out layer-wise relevance propagation (LRP) so that the salient time step can be identified. We chose a selected number of such important time indices to create the bounding region of the shortest possible classification timeframe. We identified the period 21st April 2019 to 9th August 2019 as having the best trade-off in terms of accuracy and earliness. This timeframe only suffers a 0.75% loss in accuracy as compared to using the full timeseries. We observed that the LRP-derived important timesteps also highlight small details in input values that differentiates between different classes and
format Preprint
id arxiv_https___arxiv_org_abs_2310_06574
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle XAI for Early Crop Classification
Chan, Ayshah
Schneider, Maja
Körner, Marco
Machine Learning
Applications
We propose an approach for early crop classification through identifying important timesteps with eXplainable AI (XAI) methods. Our approach consists of training a baseline crop classification model to carry out layer-wise relevance propagation (LRP) so that the salient time step can be identified. We chose a selected number of such important time indices to create the bounding region of the shortest possible classification timeframe. We identified the period 21st April 2019 to 9th August 2019 as having the best trade-off in terms of accuracy and earliness. This timeframe only suffers a 0.75% loss in accuracy as compared to using the full timeseries. We observed that the LRP-derived important timesteps also highlight small details in input values that differentiates between different classes and
title XAI for Early Crop Classification
topic Machine Learning
Applications
url https://arxiv.org/abs/2310.06574