Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/122466
Title: Predicting human behaviour in public good games using a game theoretical approach
Author: Vilarasau Antolin, Ignasi
Tutor: Vicens Bennasar, Julian Antonio
Abstract: Social interactions are present on every daily situation, and social situations that involve strategic behaviour are the keys of every social interaction. Those situations can be extrapolated to Evolutionary Game Theory through Public Goods Games. On this sense, studying the behaviour of individuals from the theoretical and experimental point of view can lead to an understanding how individuals would react and act to those strategic situations and it could be very important in order to be able to anticipate the outcomes of every kind of conflict or social situation. In order to be able to understand that behaviour we will try to identify discrete behavioural types (clusters) of individuals in experimental data and try to classify each individual behaviour to one of the types of discrete behaviours identified. In addition, we will also apply more machine learning supervised models in order to be able to classify the individuals with the maximum accuracy. Speaking of machine learning models allows us to also speak about eXplainable Artificial Intelligence (XAI). XAI creates a suite of machine learning techniques that enables humans to understand, appropriately trust, and effectively manage the emerging generation of artificially intelligent partners (5). So, in addition to applying all the machine learning models that we previously talked about, we will also introduce the main aspects of XAI techniques/methods and apply them to our ML models. Thus, we will finally apply a set of algorithms, based on game theory, that corresponds to the contribution of each feature of the model towards pushing the prediction away from the expected value.
Keywords: public goods games
machine learning
explainability
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Sep-2020
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
ivilarasau0TFM0920memory.pdfMemory of TFM1,28 MBAdobe PDFThumbnail
View/Open