Introductory comments by the organisers.

Human decision-making is often described as irrational, being the result of applying error-prone heuristics. I will argue that this is partly a consequence of the use of an unrealistic standard of rationality, and that the notion of bounded optimality from the artificial intelligence literature provides a better framework for understanding human behaviour. Within this framework a rational agent seeks to execute the best algorithm for solving a problem, taking into account available computational resources and the cost of time. We find that several classic heuristics from the decision-making literature are bounded optimal, assuming people have access to particular computational resources. This establishes a new problem: how do people find such good heuristics? I will discuss how this problem can be addressed via rational metareasoning, which examines how rational agents should decide what algorithm to use in solving a problem. The result is a view of human decision-making in which people are intelligently and flexibly making the most of their limited computational resources.

It is argued that, contrary to a rather prevalent view within economic theory, rationality does not imply Bayesianism. The note begins by defining these terms and justifying the choice of these definitions, proceeds to survey the main justification for this prevalent view, and concludes by highlighting its weaknesses.

We study an information-theoretic framework of bounded rational decision-making that trades off utility maximization against information-processing costs. We apply the basic principle of this framework to perception-action systems and show how the formation of abstractions and decision-making hierarchies depends on information-processing costs.

Posters: Marcus Buckmann, Özgür Simsek: Decision Heuristics For Comparison: How Good Are They? Sam Ganzfried: Optimal Number of Choices in Rating Contexts. Jan Malte Lichtenberg, Özgür Simsek: Simple Regression Models. Miroslav Kárný: Towards Implementable Prescriptive Decision Making. Krzysztof Drachal: Forecasting Spot Oil Price Using Google Probabilities. Özgür Simsek, Marcus Buckmann: On Learning Decision Heuristics. Marko Ruman, František Hula, Tatiana V. Guy, Miroslav Kárný: Real-Life Performance of Deliberation-Aware Responder in Multi-Proposer Ultimatum Game. Shaudi Mahdavi, Mohammad Amin Rahimian: Does Hindsight Bias Impede Learning? Vladimíra Seckarová: Performance of Kullback-Leibler Based Expert Opinion Pooling for Unlikely Events. Jakub Štech, T.V.Guy: Lazy-learning fully probabilistic decision making. Dimitri Ognibene, Vincenzo G. Fiore, Xiaosi Gu: Addiction in a Bounded Rational Model: the role of Exploration and Environment Structure.

For planning with high uncertainty, or with too many possible end positions as in games like Go or even chess, one can almost never solve the optimal control problem and must use some receding horizon heuristics. One such heuristics is based on the idea of maximizing empowerment, namely, keep the number of possible options maximal. This has been formulated using information theoretic ideas as maximizing the information capacity between the sequence of actions and the possible state of the system at some finite horizon, but no efficient algorithm for calculating this capacity was suggested. In this work we propose a concrete and efficient way for calculating the capacity between a sequence of actions and future states, based on local linearization of the dynamics and Gaussian channel capacity calculation. I will describe the new algorithm and some of its interesting implications.

The purpose of this note is to outline a framework for uncertain reasoning which drops unrealistic assumptions about the agents' inferential capabilities. To do so, we envisage a pivotal role for the recent research programme of depth-bounded Boolean logics (D'Agostino et al., 2013). We suggest that this can be fruitfully extended to the representation of rational belief under uncertainty. By doing this we lay the foundations for a prescriptive account of rational belief, namely one that realistic agents, as opposed to idealised ones, can feasibly act upon.

Individuals are often faced with temptations that can lead them astray from long-term goals. We’re interested in developing interventions that steer individuals toward making good initial decisions and then maintaining those decisions over time. In the realm of financial decision making, a particularly successful approach is the prize-linked savings account: individuals are incentivized to make deposits by tying deposits to a periodic lottery that awards bonuses to the savers. Although these lotteries have been very effective in motivating savers across the globe, they are a one-size-fits-all solution. We investigate whether customized bonuses can be more effective. We formalize a delayed-gratification task as a Markov decision problem and characterize individuals as rational agents subject to temporal discounting, costs associated with effort, and moment-to-moment fluctuations in willpower. Our theory is able to explain key behavioral findings in intertemporal choice. We created an online delayed-gratification game in which the player scores points by choosing a queue to wait in and patiently advancing to the front. Data collected from the game is fit to the model, and the instantiated model is then used to optimize predicted player performance over a space of incentives. We demonstrate that customized incentive structures can improve goal-directed decision making.

Quantum probability theory (QPT) is a probabilistic framework, alternative to Classic Probability Theory (CPT) that has been employed to model some of the paradoxical phenomena found with human judgments and decisions. One question that arises, however, is why an agent might behave this way especially given that these judgments and decisions appear to deviate from rationality? We will argue that QPT can fulfill the requirement for the Dutch Book theorem, which has been used to justify the rational status of CPT. A second question is how these quantum processes work? We will show how the heuristic processes people use to make judgments and decisions can be modeled with quantum information theory, which perhaps paradoxically provides a better and more parsimonious description of these boundedly rational heuristic processes people use than models grounded in classic information theory. In sum, we will argue that QPT can offer a principled account of the processes people use to make judgments and decisions with their limited computational resources and those judgments and decisions can nevertheless be quite rational.

We develop a theory of quantum rational decision making in the tradition of Anscombe and Aumann’s axiomatisation of preferences on horse lotteries. It is essentially the Bayesian decision theory generalised to the space of Hermitian matrices. Among other things, this leads us to give a representation theorem showing that quantum complete rational preferences are obtained by means of expected utility considerations.

We formalize the idea of probability distributions that lead to reliable predictions about some, but not all aspects of a domain. The resulting notion of safety' provides a fresh perspective on foundational issues in statistics, providing a middle ground between imprecise probability and multiple-prior models on the one hand and strictly Bayesian approaches on the other. It also allows us to formalize fiducial distributions in terms of the set of random variables that they can safely predict, thus taking some of the sting out of the fiducial idea. By restricting probabilistic inference to safe uses, one also automatically avoids paradoxes such as the Monty Hall problem. Safety comes in a variety of degrees, such asvalidity' (the strongest notion), calibration',confidence safety' and `unbiasedness' (almost the weakest notion).

We review the distinction between evidential and causal decision-making and the challenges that this distinction poses to the application of the expected utility principle. We furthermore establish firm connections between causality, information-theory, and game-theoretic concepts. Finally, we show how to use the aforementioned connections to construct adaptive agents that are universal over a given class of stochastic environments - such as Thompson sampling.