Decision Theory

Recursive expected utility and the separation of attitudes towards risk and ambiguity: an experimental study

by Sujoy Chakravarty

Are You Risk Averse Over Other People's Money?

by Sujoy Chakravarty

Newcomb’s problem: the causalists get rich

by Phyllis Illari

Analysis 64.2, April 2004, pp. 187–89
doi:10.1093/analys/64.2.187

Freedom, Fiction and Evidential Decision Theory

by Phyllis Illari

Erkenntnis (2007) 66:393--407

This paper argues against evidential decision-theory, by showing that the newest responses to its biggest current... more This paper argues against evidential decision-theory, by showing that the newest responses to its biggest current problem – the medical Newcomb problems – don’t work. The latest approach is described, and the arguments of two main proponents of it – Huw Price and CR Hitchcock – examined. It is argued that since neither new defence is successful, causation remains essential to understanding means-end agency.

Abel, D. L. , 2008, 'The Cybernetic Cut': Progressing from description to prescription in systems theory, The Open Cybernetics and Systemics Journal, 2, 234-244

by Dr. David L. Abel

See aso: Abel, D.L., The Cybernetic Cut [Scirus Topic Page].

http://www.scitopics.com/The_Cybernetic_Cut.html

(Last accessed January, 2012).

Howard Pattee championed the term “epistemic cut” to describe the symbol-matter, subject-object, genotype-phenotype... more Howard Pattee championed the term “epistemic cut” to describe the symbol-matter, subject-object, genotype-phenotype distinction. But the precise point of contact between formalism and physicality still needs elucidation. Can information be physical? How does nonphysical mind arise from physicality to then establish formal control over that physicality (e.g., engineering feats, computer science)? How did inanimate nature give rise to an algorithmically organized, semiotic and cybernetic life? Both the practice of physics and life itself require traversing not only an epistemic cut, but a Cybernetic Cut. A fundamental dichotomy of reality is delineated. The dynamics of physicality (“chance and necessity”) lie on one side. On the other side lies the ability to choose with intent what aspects of ontological being will be preferred, pursued, selected, rearranged, integrated, organized, preserved, and used (formalism).

The Story of Rational Action

by J. David Velleman

Originally published in Philosophical Topics (1993). Reprinted in The Possibility of Practical Reason, online in the SPO Monograph Series.

An explanation for the normativity of the axioms of decision theory. An explanation for the normativity of the axioms of decision theory.

Precautionary Principle

by Christian Munthe

Accepted for publication and forthcoming in LaFolette, H, Deigh, J & Stroud, S (ed.), The International Encyclopedia of Ethics, Wiley Blackwell, forthcoming.

Abel, D.L.; Trevors, J.T., 2006, Self-Organization vs. Self-Ordering events in life-origin models, Physics of Life Reviews, 3, 211-228.

by Dr. David L. Abel

Self-ordering phenomena should not be confused with self-organization.  Self-ordering events occur spontaneously... more Self-ordering phenomena should not be confused with self-organization.  Self-ordering events occur spontaneously according to natural “law” propensities and are purely physicodynamic.  Crystallization and the spontaneously forming dissipative structures of Prigogine are examples of self-ordering.  Self-ordering phenomena involve no decision nodes, no dynamically-inert configurable switches, no logic gates, no steering toward algorithmic success or “computational halting.”  Hypercycles, genetic and evolutionary algorithms, neural nets, and cellular automata have not been shown to self-organize spontaneously into nontrivial functions.  Laws and fractals are both compression algorithms containing minimal complexity and information.  Organization typically contains large quantities of prescriptive information.  Prescriptive information either instructs or directly produces nontrivial optimized algorithmic function at its destination.  Prescription requires choice contingency rather than chance contingency or necessity.  Organization requires prescription, and is abstract, conceptual, formal, and algorithmic.  Organization utilizes a sign/symbol/token system to represent many configurable switch settings.  Physical switch settings allow instantiation of nonphysical selections for function into physicality.  Switch settings represent choices at successive decision nodes that integrate circuits and instantiate cooperative management into conceptual physical systems.  Switch positions must be freely selectable to function as logic gates.  Switches must be set according to rules, not laws.  Inanimacy cannot “organize” itself.  Inanimacy can only self-order.  “Self-organization” is without empirical and prediction-fulfilling support.  No falsifiable theory of self-organization exists.  “Self-organization” provides no mechanism and offers no detailed verifiable explanatory power.  Care should be taken not to use the term “self-organization” erroneously to refer to low-informational, natural-process, self-ordering events, especially when discussing genetic information.

Abel, D.L., 2009, The biosemiosis of prescriptive information Semiotica, 2009, (174) 1-19

by Dr. David L. Abel

Nothing will ever be accomplished in discussing "information" and its source in biology until we narrow down the different types of information. Functional Information (FI) has two subsets: Descriptive Informtion (DI) and Prescriptive Information (PI). Genomic and Epigenomic information is Prescripive Information (PI). This paper is about PI and is role in the cybernetics (metaboli controls, especially) of all known life.

Prescriptive information either instructs or directly produces nontrivial function at its destination (Abel and Trevors, 2005; Abel and Trevors, 2006a). Prescriptive information (PI) does far more than describe. As its name implies, PI specifically conceives and prescribes utility. PI either tells us what choices to make, or it is a recordation of wise choices already made (Abel and Trevors, 2007). When we buy computer software, we are purchasing PI. PI can extend beyond instruction into the realization of non-trivial “halting” cybernetic function. It can perform nonphysical “formal work.” PI can then be instantiated into physicality to marshal physical work out of formal work. Cybernetic programming is only one of many forms of PI. Ordinary language itself, various communicative symbol systems, logic theory, mathematics, rules of any kind, and all types of controlling and computational algorithms are forms of PI.

PI arises from expedient choice commitments at bona fide decision nodes (Abel and Trevors, 2006b; Kaplan, 1996). Such decisions steer events toward pragmatic results that are valued by agents. Empirical evidence of PI arising spontaneously from inanimate nature is sorely lacking (Abel and Trevors, 2006b). Neither chance nor necessity has been shown to generate prescriptive information (Trevors and Abel, 2004). Choice contingency, not chance contingency, prescribes non-trivial function.

See also: Abel, D.L., Prescriptive Information (PI) [Scirus Topic Page]

http://www.scitopics.com/Prescriptive_Information_PI.html

(Last accessed January, 2012).

Exactly how do the sign/symbol/token systems of endo- and exo-biosemiosis differ from those of cognitive semiosis? Do... more Exactly how do the sign/symbol/token systems of endo- and exo-biosemiosis differ from those of cognitive semiosis? Do the biological messages that integrate metabolism have conceptual meaning? Semantic information has two subsets: Descriptive and Prescriptive. Prescriptive information instructs or directly produces nontrivial function. In cognitive semiosis, prescriptive information requires anticipation and “choice with intent” at bona fide decision nodes. Prescriptive information either tells us what choices to make, or it is a recordation of wise choices already made. Symbol systems allow recordation of deliberate choices and the transmission of linear digital prescriptive information. Formal symbol selection can be instantiated into physicality using physical symbol vehicles (tokens). Material symbol systems (MSS) formally assign representational meaning to physical objects. Even verbal semiosis instantiates meaning into physical sound waves using an MSS. Formal function can also be incorporated into physicality through the use of dynamically-inert (dynamically-incoherent or -decoupled) configurable switch-settings in conceptual circuits. This paper examines the degree to which biosemiosis conforms to the essential formal criteria of prescriptive semiosis and cybernetic management.

Keywords: Bifurcation points; Chance contingency; Choice contingency; Complexity; Computation; Configurable switches; Cybernetics; The Cybernetic Cut; Decision nodes; Decision theory; Dynamics; Emergence; Formalism; Materialism; Material Symbol Systems (MSS); Physicalism; Physical symbol vehicles; Sign systems; Symbol systems; Systems theory; Tokens.

Diachronic Rationality and Prediction-Based Games

by David Wallace

Proceedings of the Aristotelian Society 110 (2010) pp. 243-266

I explore the debate about causal versus evidential decision theory, and its recent developments in the work of Andy... more I explore the debate about causal versus evidential decision theory, and its recent developments in the work of Andy Egan, through the method of some simple games based on agents' predictions of each other's actions. My main focus is on the requirement for rational agents to act in a way which is consistent over time and its implications for such games and their more realistic cousins.

'Savage vs. Wald: Was Bayesian Decision Theory the Only Available Alternative for Postwar Economics?

by Nicola Giocoli

The paper compares the two main approaches to decision-making under uncertainty in the early 1950s, Wald's minimax... more The paper compares the two main approaches to decision-making under uncertainty in the early 1950s, Wald's minimax theory (Wald 1950) and Savage's subjective expected utility theory (Savage 1954), from the viewpoint of postwar neoclassical economics. It is argued that, while obvious reasons led to the success of Savage's approach - the main one being of course its dependence upon the familiar notion of utility maximization - the latter also entailed a significant change in the traditional neoclassical representation of rational behavior, in particular as far as the role of subjectivism and of consistency restrictions were concerned. Despite its severe limitations, Wald's approach was instead immune from these shortcomings and pointed at an "objectivization" of decision-making.

Efficient Dominance Testing for Unconditional Preferences

by Vasant Honavar

Santhanam, G., Basu, S., and Honavar, V. (2010). Efficient Dominance Testing for Unconditional Preferences. In: Proceedings of the Twelfth International Conference on the Principles of Knowledge Representation and Reasoning (KR 2010). pp. 590-592. AAAI Press.

We study a dominance relation for comparing outcomes based on unconditional qualitative preferences and compare it... more We study a dominance relation for comparing outcomes based on unconditional qualitative preferences and compare it with its unconditional counterparts for TCP-nets and their variants. Dominance testing based on this relation can be car- ried out in polynomial time by evaluating the satisfiability of a logic formula.

Ambivalent Desires and the Problem with Reduction

by Derek Baker

'Philosophical Studies' vol. 150, no. 1, pp. 37-47, August 2010.

Ambivalence is most naturally characterized as a case of conflicting desires. In most cases, an agent’s intrinsic... more Ambivalence is most naturally characterized as a case of conflicting desires. In most cases, an agent’s intrinsic desires conflict contingently: there is some possible world in which both desires would be satisfied. This paper argues, though, that there are cases in which intrinsic desires necessarily conflict—i.e., the desires are not jointly satisfiable in any possible world. Desiring a challenge for its own sake is a paradigm case of such a desire. Ambivalence of this sort in an agent’s desires creates special problems for the project of reducing all facts about an agent’s desires to facts about his or her preferences over options. If this reductive project fails, there is reason to suspect that the Decision Theory cannot give us a complete theory of Humean rationality.

Moral Uncertainty and Intertheoretic Comparisons of Value

by William Crouch

My BPhil Thesis, 2010.

Prioritizing conservation investments for mammal species globally

by Megan Evans

Kerrie A. Wilson1,*, Megan C. Evans1, Moreno Di Marco3, David C. Green2, Luigi Boitani3, Hugh P. Possingham1, Federica Chiozza3 and Carlo Rondinini3

We need to set priorities for conservation because we cannot do everything, everywhere, at the same time. We... more We need to set priorities for conservation because we cannot do everything, everywhere, at the same time. We determined priority areas for investment in threat abatement actions, in both a cost-effective and spatially and temporally explicit way, for the threatened mammals of the world. Our analysis presents the first fine-resolution prioritization analysis for mammals at a global scale that accounts for the risk of habitat loss, the actions required to abate this risk, the costs of these actions and the likelihood of investment success. We evaluated the likelihood of success of investments using information on the past frequency and duration of legislative effectiveness at a country scale. The establishment of new protected areas was the action receiving the greatest investment, while restoration was never chosen. The resolution of the analysis and the incorporation of likelihood of success made little difference to this result, but affected the spatial location of these investments.

Decision-making under great uncertainty: environmental management in an era of global change

by Sturle Hauge Simonsen

Polasky, S., S.R. Carpenter, C. Folke, and B. Keeler. 2011. Decision-Making under Great Uncertainty: Environmental Management in an Era of Global Change. Trends in Ecology and Evolution doi:10.1016/j.tree.2011.04.007.

Global change issues are complex and the consequences of decisions are often highly uncertain. The large spatial and... more Global change issues are complex and the consequences of decisions are often highly uncertain. The large spatial and temporal scales and stakes involved make it important to take account of present and potential consequences in decision-making.

Standard approaches to decision-making under uncertainty require information about the likelihood of alternative states, how states and actions combine to form outcomes and the net benefits of different outcomes.

For global change issues, however, the set of potential states is often unknown, much less the probabilities, effect of actions or their net benefits.

Decision theory, thresholds, scenarios and resilience thinking can expand awareness of the potential states and outcomes, as well as of the probabilities and consequences of outcomes under alternative decisions.

Agency, Determinism, Focal Time Frames, and Narrative in Processive Minimalist Music

by Joshua B. Mailman

To appear as Chapter 6 of the book:
Music and Narrative Since 1900.
Michael Klein and Nicholas Reyland, (Eds.) Indiana University Press. Forthcoming in November 2012.
http://www.iupress.indiana.edu/product_info.php?cPath=1037_3025_3988&products_id=806644

The Suppositional Ramsey Test and Decision-Instability

by Simone Duca

Implementing a Decision-Theoretic Design In Clinical Trials: Why and How?

by Harutyun Shahumyan

Palmer CR, Shahumyan H.
Centre for Applied Medical Statistics, University of Cambridge, Cambridge, UK.

Statistics in Medicine

Special Issue: French Society of Statistics, Biopharmacy and Health Group Fifth International Meeting on Statistical Methods in Biopharmacy “Statistical innovations in clinical trials”

Volume 26, Issue 27, pages 4939–4957, 30 November 2007

This paper addresses two main questions: first, why should Bayesian and other innovative, data-dependent design models... more This paper addresses two main questions: first, why should Bayesian and other innovative, data-dependent design models be put into practice and, secondly, given the past dearth of actual applications, how might one example of such a design be implemented in a genuine example trial?

Clinical trials amalgamate theory, practice and ethics, but this last point has become relegated to the background, rather than taking often a more appropriate primary role. Trial practice has evolved but has its roots in R. A. Fisher's randomized agricultural field trials of the 1920s. Reasons for, and consequences of, this are discussed from an ethical standpoint, drawing on an under-used dichotomy introduced by French authors Lellouch and Schwartz (Int. Statist. Rev. 1971; 39:27–36). Plenty of ethically motivated designs for trials, including Bayesian designs have been proposed, but have found little application thus far. One reason for this is a lack of awareness of such alternative designs among trialists, while another reason is a lack of user-friendly software to allow study simulations.

To encourage implementation, a new C++ program called ‘Daniel’ is introduced, offering much potential to assist the design of today's randomized controlled trials. Daniel evaluates a particular decision-theoretic method suitable for coping with either two or three Bernoulli response treatments with input features allowing user-specified choices of: patient horizon (number to be treated before and after the comparative stages of the trial); an arbitrary fixed trial truncation size (to allow ready comparison with traditional designs or to cope with practical constraints); anticipated success rates and a measure of their uncertainty (a matter ignored in standard power calculations); and clinically relevant, and irrelevant, differences in treatment effect sizes. Error probabilities and expected trial durations can be thoroughly explored via simulation, it being better by far to harm ‘computer patients’ instead of real ones.

Suppose the objective in a clinical trial is to select between two treatments using a maximum horizon of 500 patients, when the truly superior treatment is expected to yield a 40 per cent success rate, but is believed to really range between 20 and 60 per cent. Simulation studies show that to detect a clinically relevant, absolute difference of 10 per cent between treatments, simulation studies show the decision-theoretic procedure would treat a mean 68 pairs of patients (SD 37) before correctly identifying the better treatment 96.7 per cent of the time, an error rate of 3.3 per cent. Having made a recommendation based on these patients, the remaining, on average 364 individuals, could either be given the indicated treatment, knowing its choice is optimal for the chosen horizon, or, alternatively, they could be entered into another, separate clinical trial. For comparison, a fixed sample size trial, with standard 5 per cent level of significance and 80 per cent power to detect a 10 per cent difference, requires treating over 700 patients in two groups, with the half allocated to inferior treatment considerably outnumbering the 68 expected under the decision-theoretic design, and the overall number simply too high for realistic application.

In brief, the keys to answering the above ‘why?’ and ‘how?’ questions are ethics and software, respectively. Wider implications, both pros and cons, of implementing the particular method described will be discussed, with the overall conclusion that, where appropriate, clinical trials are now ready to undergo modernization from the agricultural age to the information age. Copyright © 2007 John Wiley & Sons, Ltd.