Improving Human Decision-Making using Metalevel-RL and Bayesian Inference
Teaching clever heuristics is a promising approach to improve decision-making. We can leverage machine learning to discover clever strategies automatically. Current methods require an accurate model of the decision problems people face in real life. But most models are misspecified because of limited in-formation and cognitive biases. To address this problem we develop strategy discovery methods that are robust to model misspecification. Robustness is achieved by modeling model-misspecification using common cognitive biases and handling uncertainty about the real-world according to Bayesian inference. We translate our methods into an intelligent tutor that automatically discovers and teaches robust planning strategies. Our robust cognitive tutor significantly improved human decision-making when the model was so biased that conventional cognitive tutors were no longer effective. These findings highlight that our robust strategy discovery methods are a significant step towards leveraging artificial intelligence to improve human decision-making in the real world.Teaching clever heuristics is a promising approach to improve decision-making. We can leverage machine learning to dis- cover clever strategies automatically. Current methods require an accurate model of the decision problems people face in real life. But most models are misspecified because of limited information and cognitive biases. To address this problem we develop strategy discovery methods that are robust to model misspecification. Robustness is achieved by modeling model-misspecification using common cognitive biases and handling uncertainty about the real-world according to Bayesian inference. We translate our methods into an intelligent tutor that automatically discovers and teaches robust planning strategies. Our robust cognitive tutor significantly improved human decision-making when the model was so biased that conventional cognitive tutors were no longer effective. These findings highlight that our robust strategy discovery methods are a significant step towards leveraging artificial intelligence to improve human decision-making in the real world.
| Author(s): | Anirudha Kemtur and Yash Raj Jain and Aashay Mehta and Frederick Callaway and Saksham Consul and Jugoslav Stojcheski and Falk Lieder |
| Year: | 2020 |
| Month: | December |
| Bibtex Type: | Article (article) |
| State: | Accepted |
| Electronic Archiving: | grant_archive |
| Event Name: | NeurIPS Workshop on Challenges for Real-World RL |
| Links: | |
BibTex
@article{Kemtur2020NeurIPS-Decision,
title = {Improving Human Decision-Making using Metalevel-RL and Bayesian Inference},
abstract = {Teaching clever heuristics is a promising approach to improve decision-making. We can leverage machine learning to discover clever strategies automatically. Current methods require an accurate model of the decision problems people face in real life. But most models are misspecified because of limited in-formation and cognitive biases. To address this problem we
develop strategy discovery methods that are robust to model misspecification. Robustness is achieved by modeling model-misspecification using common cognitive biases and handling uncertainty about the real-world according to Bayesian inference. We translate our methods into an intelligent tutor that
automatically discovers and teaches robust planning strategies. Our robust cognitive tutor significantly improved human decision-making when the model was so biased that conventional cognitive tutors were no longer effective. These findings highlight that our robust strategy discovery methods are a significant step towards leveraging artificial intelligence to improve human decision-making in the real world.Teaching clever heuristics is a promising approach to improve decision-making. We can leverage machine learning to dis-
cover clever strategies automatically. Current methods require an accurate model of the decision problems people face in real life. But most models are misspecified because of limited information and cognitive biases. To address this problem we develop strategy discovery methods that are robust to model misspecification. Robustness is achieved by modeling model-misspecification using common cognitive biases and handling uncertainty about the real-world according to Bayesian inference. We translate our methods into an intelligent tutor that automatically discovers and teaches robust planning strategies. Our robust cognitive tutor significantly improved human decision-making when the model was so biased that conventional cognitive tutors were no longer effective. These findings highlight that our robust strategy discovery methods are a significant step towards leveraging artificial intelligence to
improve human decision-making in the real world.},
month = dec,
year = {2020},
slug = {kemtur2020neurips-decision},
author = {Kemtur, Anirudha and Jain, Yash Raj and Mehta, Aashay and Callaway, Frederick and Consul, Saksham and Stojcheski, Jugoslav and Lieder, Falk},
month_numeric = {12}
}