Human-inspired strategies to solve complex joint tasks in multi agent systems

Auletta, Fabrizia; Bernardo, Mario di; Richardson, Michael J.

doi:10.1016/j.ifacol.2021.11.033

Cited by 7 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 – 4 , LSTM NN models trained using sequence lengths of 0.5 to 2 s could accurately predict (above 95%) target selection decision at prediction horizons ranging from 20 ms to 2.56 s. Moreover, although correct predictions at T hor ≥ 16 does not provide definitive evidence that these predictions preceded a herders intent, this possibility seems likely as the action decisions made by human actors during skillful action are spontaneously tuned responses to the unfolding dynamics of a task 13 , 18 , 20 and, for the type of perceptual-motor task investigated here, often only occur 150 to 300 ms prior to action onset 78 . A significant implication is that the current modeling approach could be employed for the anticipatory correction of human action decisions during task training and real-time task engagement, as well as to develop more ‘human-like’ artificial or robotic agents 79 that are capable of robustly forecasting and reciprocally adjusting to the behavior of human co-actors during human–machine interaction contexts.…”

Section: Discussionmentioning

confidence: 99%

Predicting and understanding human action decisions during skillful joint-action using supervised machine learning and explainable-AI

Auletta

Kallen

Bernardo

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

This study investigated the utility of supervised machine learning (SML) and explainable artificial intelligence (AI) techniques for modeling and understanding human decision-making during multiagent task performance. Long short-term memory (LSTM) networks were trained to predict the target selection decisions of expert and novice players completing a multiagent herding task. The results revealed that the trained LSTM models could not only accurately predict the target selection decisions of expert and novice players but that these predictions could be made at timescales that preceded a player’s conscious intent. Importantly, the models were also expertise specific, in that models trained to predict the target selection decisions of experts could not accurately predict the target selection decisions of novices (and vice versa). To understand what differentiated expert and novice target selection decisions, we employed the explainable-AI technique, SHapley Additive explanation (SHAP), to identify what informational features (variables) most influenced modelpredictions. The SHAP analysis revealed that experts were more reliant on information about target direction of heading and the location of coherders (i.e., other players) compared to novices. The implications and assumptions underlying the use of SML and explainable-AI techniques for investigating and understanding human decision-making are discussed.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting and understanding human action decisions during skillful joint-action using supervised machine learning and explainable-AI

Auletta

Kallen

Bernardo

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, although correct predictions at T hor ≥ 16 does not provide definitive evidence that these predictions preceded a herders intent, this possibility seems likely as the action decisions made by human actors during skillful action are spontaneously tuned responses to the unfolding dynamics of a task [14,64,30] and, for the type of perceptual-motor task investigated here, often only occur 150 to 300 ms prior to action onset [70]. A significant implication is that the current modeling approach could be employed for the anticipatory correction of human action decisions during task training and real-time task engagement, as well as to develop more 'human-like' artificial or robotic agents ( [6]) that are capable of robustly forecasting and reciprocally adjusting to the behavior of human co-actors within human-machine interaction contexts.…”

Section: Discussionmentioning

confidence: 99%

Predicting and Understanding Human Action Decisions during Skillful Joint-Action via Machine Learning and Explainable-AI

Auletta¹,

Kallen²,

Bernardo³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This study uses supervised machine learning (SML) and explainable artificial intelligence (AI) to model, predict and understand human decision-making during skillful joint-action. Long short-term memory networks were trained to predict the target selection decisions of expert and novice actors completing a dyadic herding task. Results revealed that the trained models were expertise specific and could not only accurately predict the target selection decisions of expert and novice herders but could do so at timescales that preceded an actor's conscious intent. To understand what differentiated the target selection decisions of expert and novice actors, we then employed the explainable-AI technique, SHapley Additive exPlanation, to identify the importance of informational features (variables) on model predictions. This analysis revealed that experts were more influenced by information about the state of their co-herders compared to novices. The utility of employing SML and explainable-AI techniques for investigating human decision-making is discussed.

show abstract

“…Moreover, although correct predictions at T hor ≥ 16 does not provide definitive evidence that these predictions preceded a herders intent, this possibility seems likely as the action decisions made by human actors during skillful action are spontaneously tuned responses to the unfolding dynamics of a task [13,18,20] and, for the type of perceptual-motor task investigated here, often only occur 150 to 300 ms prior to action onset [70]. A significant implication is that the current modeling approach could be employed for the anticipatory correction of human action decisions during task training and real-time task engagement, as well as to develop more 'human-like' artificial or robotic agents ( [71]) that are capable of robustly forecasting and reciprocally adjusting to the behavior of human co-actors within human-machine interaction contexts.…”

Section: Discussionmentioning

confidence: 99%

Predicting and Understanding Human Action Decisions during Skillful Joint-Action using Supervised Machine Learning and Explainable-AI

Auletta

Kallen

Bernardo

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This study investigated the utility of supervised machine learning (SML) and explainable artificial intelligence (AI) techniques for modeling and understanding human decision-making during multiagent task performance. Long short-term memory (LSTM) networks were trained to predict the target selection decisions of expert and novice players completing a multiagent herding task. The results revealed that the trained LSTM models could not only accurately predict the target selection decisions of expert and novice players but that these predictions could be made at timescales that preceded a player’s conscious intent. Importantly, the models were also expertise specific, in that models trained to predict the target selection decisions of experts could not accurately predict the target selection decisions of novices (and vice versa). To understand what differentiated expert and novice target selection decisions, we employed the explainable-AI technique, SHapley Additive explanation (SHAP), to identify what informational features (variables) most influenced model predictions. The SHAP analysis revealed that experts were more reliant on information about target direction of heading and the location of co-herders (i.e., other players) compared to novices. The implications and assumptions underlying the use of SML and explainable-AI techniques for investigating and understanding human decision-making are discussed.

show abstract

Human-inspired strategies to solve complex joint tasks in multi agent systems

Cited by 7 publications

References 12 publications

Predicting and understanding human action decisions during skillful joint-action using supervised machine learning and explainable-AI

Predicting and understanding human action decisions during skillful joint-action using supervised machine learning and explainable-AI

Predicting and Understanding Human Action Decisions during Skillful Joint-Action via Machine Learning and Explainable-AI

Predicting and Understanding Human Action Decisions during Skillful Joint-Action using Supervised Machine Learning and Explainable-AI

Contact Info

Product

Resources

About