Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games
We present results of attempts to expand and enhance the predictive power of Early Warning Signals (EWS) for Critical Transitions (Scheffer et al. 2009) through the deployment of a Long-Short-Term-Memory (LSTM) Neural Network on agent-based simulations of a Repeated Public Good Game, which due to po...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-01-01
|
| Series: | MethodsX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2215016120301394 |
| id |
doaj-80ccd478ae6e4c17b2a6e35791048717 |
|---|---|
| record_format |
Article |
| spelling |
doaj-80ccd478ae6e4c17b2a6e357910487172021-01-02T05:10:29ZengElsevierMethodsX2215-01612020-01-017100920Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good gamesManfred Füllsack0Marie Kapeller1Simon Plakolb2Georg Jäger3Corresponding author.; Institute of Systems Sciences, Innovation and Sustainability Research, University of Graz, Graz, AustriaInstitute of Systems Sciences, Innovation and Sustainability Research, University of Graz, Graz, AustriaInstitute of Systems Sciences, Innovation and Sustainability Research, University of Graz, Graz, AustriaInstitute of Systems Sciences, Innovation and Sustainability Research, University of Graz, Graz, AustriaWe present results of attempts to expand and enhance the predictive power of Early Warning Signals (EWS) for Critical Transitions (Scheffer et al. 2009) through the deployment of a Long-Short-Term-Memory (LSTM) Neural Network on agent-based simulations of a Repeated Public Good Game, which due to positive feedbacks on experience and social entrainment transits abruptly from majority cooperation to majority defection and back. Our method extension is inspired by several known deficiencies of EWS and by lacking possibilities to consider micro-level interaction in the so far primarily used simulation methods. We find that • The method is applicable to agent-based simulations (as an extension of equation-based methods). • The LSTM yields signals of imminent transitions that can complement statistical indicators of EWS. • The less tensely connected part of an agent population could take a larger role in causing a tipping than the well-connected part.http://www.sciencedirect.com/science/article/pii/S2215016120301394Critical transitionsEarly warning signalsLong-short-term-memory neural networksAgent-based modelRepeated public good gameScale-free networks |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Manfred Füllsack Marie Kapeller Simon Plakolb Georg Jäger |
| spellingShingle |
Manfred Füllsack Marie Kapeller Simon Plakolb Georg Jäger Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games MethodsX Critical transitions Early warning signals Long-short-term-memory neural networks Agent-based model Repeated public good game Scale-free networks |
| author_facet |
Manfred Füllsack Marie Kapeller Simon Plakolb Georg Jäger |
| author_sort |
Manfred Füllsack |
| title |
Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| title_short |
Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| title_full |
Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| title_fullStr |
Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| title_full_unstemmed |
Training LSTM-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| title_sort |
training lstm-neural networks on early warning signals of declining cooperation in simulated repeated public good games |
| publisher |
Elsevier |
| series |
MethodsX |
| issn |
2215-0161 |
| publishDate |
2020-01-01 |
| description |
We present results of attempts to expand and enhance the predictive power of Early Warning Signals (EWS) for Critical Transitions (Scheffer et al. 2009) through the deployment of a Long-Short-Term-Memory (LSTM) Neural Network on agent-based simulations of a Repeated Public Good Game, which due to positive feedbacks on experience and social entrainment transits abruptly from majority cooperation to majority defection and back. Our method extension is inspired by several known deficiencies of EWS and by lacking possibilities to consider micro-level interaction in the so far primarily used simulation methods. We find that • The method is applicable to agent-based simulations (as an extension of equation-based methods). • The LSTM yields signals of imminent transitions that can complement statistical indicators of EWS. • The less tensely connected part of an agent population could take a larger role in causing a tipping than the well-connected part. |
| topic |
Critical transitions Early warning signals Long-short-term-memory neural networks Agent-based model Repeated public good game Scale-free networks |
| url |
http://www.sciencedirect.com/science/article/pii/S2215016120301394 |
| work_keys_str_mv |
AT manfredfullsack traininglstmneuralnetworksonearlywarningsignalsofdecliningcooperationinsimulatedrepeatedpublicgoodgames AT mariekapeller traininglstmneuralnetworksonearlywarningsignalsofdecliningcooperationinsimulatedrepeatedpublicgoodgames AT simonplakolb traininglstmneuralnetworksonearlywarningsignalsofdecliningcooperationinsimulatedrepeatedpublicgoodgames AT georgjager traininglstmneuralnetworksonearlywarningsignalsofdecliningcooperationinsimulatedrepeatedpublicgoodgames |
| _version_ |
1724359413754494976 |