A deep Generative Artificial Intelligence system to predict species coexistence patterns. Issue 5 (6th March 2022)
- Record Type:
- Journal Article
- Title:
- A deep Generative Artificial Intelligence system to predict species coexistence patterns. Issue 5 (6th March 2022)
- Main Title:
- A deep Generative Artificial Intelligence system to predict species coexistence patterns
- Authors:
- Hirn, Johannes
García, José Enrique
Montesinos‐Navarro, Alicia
Sánchez‐Martín, Ricardo
Sanz, Veronica
Verdú, Miguel - Abstract:
- Abstract: Predicting coexistence patterns is a current challenge to understand diversity maintenance, especially in rich communities where these patterns' complexity is magnified through indirect interactions that prevent their approximation with classical experimental approaches. We explore cutting‐edge Machine Learning techniques called Generative Artificial Intelligence (GenAI) to predict species coexistence patterns in vegetation patches, training generative adversarial networks (GAN) and variational AutoEncoders (VAE) that are then used to unravel some of the mechanisms behind community assemblage. The GAN accurately reproduces real patches' species composition and plant species' affinity to different soil types, and the VAE also reaches a high level of accuracy, above 99%. Using the artificially generated patches, we found that high‐order interactions tend to suppress the positive effects of low‐order interactions. Finally, by reconstructing successional trajectories, we could identify the pioneer species with larger potential to generate a high diversity of distinct patches in terms of species composition. Understanding the complexity of species coexistence patterns in diverse ecological communities requires new approaches beyond heuristic rules. Generative Artificial Intelligence can be a powerful tool to this end as it allows to overcome the inherent dimensionality of this challenge. Resumen: Predecir los patrones de coexistencia de las especies permite comprenderAbstract: Predicting coexistence patterns is a current challenge to understand diversity maintenance, especially in rich communities where these patterns' complexity is magnified through indirect interactions that prevent their approximation with classical experimental approaches. We explore cutting‐edge Machine Learning techniques called Generative Artificial Intelligence (GenAI) to predict species coexistence patterns in vegetation patches, training generative adversarial networks (GAN) and variational AutoEncoders (VAE) that are then used to unravel some of the mechanisms behind community assemblage. The GAN accurately reproduces real patches' species composition and plant species' affinity to different soil types, and the VAE also reaches a high level of accuracy, above 99%. Using the artificially generated patches, we found that high‐order interactions tend to suppress the positive effects of low‐order interactions. Finally, by reconstructing successional trajectories, we could identify the pioneer species with larger potential to generate a high diversity of distinct patches in terms of species composition. Understanding the complexity of species coexistence patterns in diverse ecological communities requires new approaches beyond heuristic rules. Generative Artificial Intelligence can be a powerful tool to this end as it allows to overcome the inherent dimensionality of this challenge. Resumen: Predecir los patrones de coexistencia de las especies permite comprender el mantenimiento de la diversidad, un reto especialmente difícil en comunidades ricas en las que la complejidad de estos patrones se magnifica por el efecto de las interacciones indirectas de tal manera que los enfoques experimentales clásicos son insuficientes. En este trabajo exploramos novedosas técnicas de aprendizaje profundo denominadas inteligencia artificial generativa (GenAI) para predecir patrones de coexistencia de especies en parches de vegetación, entrenando redes antagónicas generativas (GAN) y autocodificadores variacionales (VAE) que posteriormente se usan para desentrañar algunos de los mecanismos de ensamblaje de comunidades. El GAN reproduce con precisión la composición de especies de parches y la afinidad de las especies de plantas con diferentes tipos de suelo. El VAE también alcanza un alto nivel de precisión, superior al 99%. Usando los parches generados artificialmente, encontramos que las interacciones de orden superior tienden a suprimir los efectos positivos de las interacciones de orden inferior. Finalmente, al reconstruir las trayectorias sucesionales, pudimos identificar las especies pioneras con mayor potencial para generar un alto número de parches con composiciones específicas diferentes. Comprender la complejidad de los patrones de coexistencia de especies en diversas comunidades ecológicas requiere nuevos enfoques más allá de las reglas heurísticas. La Inteligencia Artificial Generativa puede ser una poderosa herramienta para este fin, ya que permite superar la dimensionalidad inherente a este desafío. … (more)
- Is Part Of:
- Methods in ecology and evolution. Volume 13:Issue 5(2022)
- Journal:
- Methods in ecology and evolution
- Issue:
- Volume 13:Issue 5(2022)
- Issue Display:
- Volume 13, Issue 5 (2022)
- Year:
- 2022
- Volume:
- 13
- Issue:
- 5
- Issue Sort Value:
- 2022-0013-0005-0000
- Page Start:
- 1052
- Page End:
- 1061
- Publication Date:
- 2022-03-06
- Subjects:
- artificial intelligence -- direct interactions -- generative adversarial networks -- indirect interactions -- species coexistence -- variational AutoEncoders
Ecology -- Periodicals
Evolution -- Periodicals
577 - Journal URLs:
- http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)2041-210X ↗
http://onlinelibrary.wiley.com/ ↗ - DOI:
- 10.1111/2041-210X.13827 ↗
- Languages:
- English
- ISSNs:
- 2041-210X
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 21369.xml