Filters








23,086 Hits in 8.4 sec

Uncertain Decisions Facilitate Better Preference Learning [article]

Cassidy Laidlaw, Stuart Russell
2021 arXiv   pre-print
To better understand preference learning in these cases, we study the setting of inverse decision theory (IDT), a previously proposed framework where a human is observed making non-sequential binary decisions  ...  Existing observational approaches for learning human preferences, such as inverse reinforcement learning, usually make strong assumptions about the observability of the human's environment.  ...  helping us improve the clarity of the paper.  ... 
arXiv:2106.10394v2 fatcat:6y76nejxbvddfkte5legwidoai

Artificial Intelligence and the Problem of Control [chapter]

Stuart Russell
2021 Perspectives on Digital Humanism  
A great deal of progress on reasoning, planning, and decision-making, as well as perception and learning, has occurred within the standard model.  ...  to humans.  ...  Assistance games are connected to inverse reinforcement learning, or IRL (Russell 1998; Ng and Russell 2000) , because the robot can learn more about human preferences from the observation of human behavior-a  ... 
doi:10.1007/978-3-030-86144-5_3 fatcat:6jinic3merae7icadhtk2gi4lm

Multi-Principal Assistance Games [article]

Arnaud Fickinger, Simon Zhuang, Dylan Hadfield-Menell, Stuart Russell
2020 arXiv   pre-print
initially uncertain about the humans payoff function.  ...  Impossibility theorems in social choice theory and voting theory can be applied to such games, suggesting that strategic behavior by the human principals may complicate the robots task in learning their  ...  A common assumption of inverse planning methods is that the robot does not influence the decision-making of the human.  ... 
arXiv:2007.09540v1 fatcat:nfzulxfh2rbktewqfvawvehip4

Muddling-Through and Deep Learning for Managing Large-Scale Uncertain Risks

Tony Cox
2019 Journal of Benefit-Cost Analysis  
Instead, he advocated incremental learning and improvement, or "muddling through," as both a positive and a normative theory of bureaucratic decision-making when costs and benefits are highly uncertain  ...  But sparse, delayed, uncertain, and incomplete feedback undermines the effectiveness of collective learning while muddling through, even if all participant incentives are aligned; it is no panacea.  ...  This paper reflects the enthusiastic comments and suggestions of organizers and participants in the SRA session. I am grateful for the opportunity to share, discuss, and improve these ideas.  ... 
doi:10.1017/bca.2019.17 fatcat:ch5f24amjnd7ndnptrwqw6zbo4

Reason, emotion and decision-making: risk and reward computation with feeling

Steven R. Quartz
2009 Trends in Cognitive Sciences  
Acknowledgements I would like to thank James Woodward and the anonymous reviewers for their insightful feedback on earlier versions of the manuscript.  ...  The author's work is partially funded by the Gordon and Betty Moore Foundation.  ...  A long tradition of research in judgment and decision making (JDM), stemming from choice or preference theory in microeconomics [1] and decision theory in philosophy [2] , suppose that uncertain decisions  ... 
doi:10.1016/j.tics.2009.02.003 pmid:19362037 fatcat:eoaiezhat5h53pdfjn5tlioena

Impossibility and Uncertainty Theorems in AI Value Alignment (or why your AGI should not have a utility function) [article]

Peter Eckersley
2019 arXiv   pre-print
We explore the alternative of using uncertain objectives, represented for instance as partially ordered preferences, or as probability distributions over total orders.  ...  We argue that this is a practical problem for any machine learning system (such as medical decision support systems or autonomous weapons) or rigidly rule-based bureaucracy that will make high stakes decisions  ...  an early poster version of this work at NIPS 2017, Daniel Ford for spotting issues and making valuable suggestions on the uncertain ordering bound.  ... 
arXiv:1901.00064v3 fatcat:ck3c4r5xxnes3npb4uvpji7mmq

Neural Basis of Strategic Decision Making

Daeyeol Lee, Hyojung Seo
2016 Trends in Neurosciences  
For example, the expected utility theory [1] and prospect theory [3] describe how uncertain outcomes of decision making can be evaluated, whereas the theories of temporal discounting specify how the value  ...  Human choice behaviors during social interactions often deviate from the predictions of game theory.  ...  This work was supported by the National Institute of Health (R01 DA029330 and R21 MH104460).  ... 
doi:10.1016/j.tins.2015.11.002 pmid:26688301 pmcid:PMC4713315 fatcat:5lwem3xpknhqxppywo5vfyw2mi

Decision field theory: A dynamic-cognitive approach to decision making in an uncertain environment

Jerome R. Busemeyer, James T. Townsend
1993 Psychological review  
The proposed theory is compared with 4 other theories of decision making under uncertainty.  ...  on preference, (e) speed-accuracy tradeoff effects in decision making, (f) the inverse relation between choice probability and decision time, (g) changes in the direction of preference under time pressure  ...  A major problem with descriptive SEU theories (e.g., prospect theory) is their inability to account for the fundamental variability of human preference (e.g., see Figure 1 ).  ... 
doi:10.1037/0033-295x.100.3.432 pmid:8356185 fatcat:o7vfgkhnqva2zkg6csz3efgl3m

Decision field theory: A dynamic-cognitive approach to decision making in an uncertain environment

Jerome R. Busemeyer, James T. Townsend
1993 Psychological review  
The proposed theory is compared with 4 other theories of decision making under uncertainty.  ...  on preference, (e) speed-accuracy tradeoff effects in decision making, (f) the inverse relation between choice probability and decision time, (g) changes in the direction of preference under time pressure  ...  A major problem with descriptive SEU theories (e.g., prospect theory) is their inability to account for the fundamental variability of human preference (e.g., see Figure 1 ).  ... 
doi:10.1037//0033-295x.100.3.432 fatcat:6ydswebrfjh4ri2e7q4yeaxmki

Exploiting risk–reward structures in decision making under uncertainty

Christina Leuker, Thorsten Pachur, Ralph Hertwig, Timothy J. Pleskac
2018 Cognition  
In subsequent decisions under uncertainty, participants often exploited the learned association by inferring probabilities from the magnitudes of the payoffs.  ...  (2) How do learned risk-reward relationships impact preferences in decision-making under uncertainty?  ...  Acknowledgements We would like to thank Jann Wäscher and Chantal Wysocki for assistance with data collection, and Susannah Goss for editing the manuscript.. Appendix A.  ... 
doi:10.1016/j.cognition.2018.02.019 pmid:29567432 fatcat:dokn5l5btjc33jbjfbdk3ij4pi

Additively Combining Utilities and Beliefs: Research Gaps and Algorithmic Developments

Anush Ghambaryan, Boris Gutkin, Vasily Klucharev, Etienne Koechlin
2021 Frontiers in Neuroscience  
Value-based decision making in complex environments, such as those with uncertain and volatile mapping of reward probabilities onto options, may engender computational strategies that are not necessarily  ...  In addition, we present computational intricacies of a recently developed model (named MIX model) representing an algorithmic implementation of the additive strategy in sequential decision-making with  ...  account of human choices in an uncertain and volatile environment of value-based decision making.  ... 
doi:10.3389/fnins.2021.704728 pmid:34658760 pmcid:PMC8517513 fatcat:xksusl4zpzasham4bk6fwwxaoq

Risk-sensitive Inverse Reinforcement Learning via Coherent Risk Models

Anirudha Majumdar, Sumeet Singh, Ajay Mandlekar, Marco Pavone
2017 Robotics: Science and Systems XIII  
The literature on Inverse Reinforcement Learning (IRL) typically assumes that humans take actions in order to minimize the expected value of a cost function, i.e., that humans are risk neutral.  ...  The resulting approach is demonstrated on a simulated driving game with ten human participants.  ...  ACKNOWLEDGMENTS The authors were partially supported by the Office of Naval Research, Science of Autonomy Program, under Contract N00014-15-1-2673, and by the Toyota Research Institute ("TRI").  ... 
doi:10.15607/rss.2017.xiii.069 dblp:conf/rss/MajumdarSMP17 fatcat:c4a5545yivbnnegqbplimt2vku

Page 4764 of Psychological Abstracts Vol. 80, Issue 11 [page]

1993 Psychological Abstracts  
on preference, (5) speed—accuracy trade-off effects in decision making, (6) the inverse relation between choice probability and decision time, (7) changes in the direction of preference under time pressure  ...  The proposed theory is compared with 4 other theories of decision making under uncertainty. —Journal abstract. 40081. Butler, Ruth.  ... 

Risk-Averse Biased Human Policies in Assistive Multi-Armed Bandit Settings [article]

Michael Koller, Timothy Patten, Markus Vincze
2021 arXiv   pre-print
To account for human biases such as the risk-aversion described in the Cumulative Prospect Theory, the setting is expanded to using observable rewards.  ...  We present an algorithm that increases the utility value of such human-robot teams. A brief evaluation indicates that arbitrary reward functions can be handled.  ...  The problem is transformed from a preference learning problem of arm choices to a preference learning of different discrete rewards.  ... 
arXiv:2104.05334v1 fatcat:nky5y6qx3vbofhfmtzlvz77rra

Page 1245 of Psychological Abstracts Vol. 48, Issue 6 [page]

1972 Psychological Abstracts  
A comparison of preferences by 30 in the 2 tetrad conditions showed that certain failures of minimum winning coalition theory in this experiment were not due to the possible preference for coalitions involving  ...  Results indicate that (a) Ss with a cooperative orientation sought significantly more in- formation from their coworkers than did competitively oriented Ss, (b) the amount of search was inversely related  ... 
« Previous Showing results 1 — 15 out of 23,086 results