Ronald P. Loui

Ronald P. Loui

This essay looks at decision-making with interval-valued probability measures. Existing decision methods have either supplemented expected utility methods with additional criteria of optimality, or have attempted to supplement the interval-valued measures. We advocate a new approach, which makes the following questions moot: 1. which additional criteria to use, and 2. how wide intervals should be. In order to implement the approach, we need more epistemological information. Such information can be generated by a rule of acceptance with a parameter that allows various attitudes toward error, or can simply be declared. In sketch, the argument is: 1. probability intervals are useful and natural in All. systems; 2. wide intervals avoid error, but are useless in some risk sensitive decision-making; 3. one may obtain narrower intervals if one is less cautious; 4. if bodies of knowledge can be ordered by their caution, one should perform the decision analysis with the acceptable body of knowledge that is the most cautious, of those that are useful. The resulting behavior differs from that of a behavioral probabilist (a Bayesian) because in the proposal, 5. intervals based on successive bodies of knowledge are not always nested; 6. if the agent uses a probability for a particular decision, she need not commit to that probability for credence or future decision; and 7. there may be no acceptable body of knowledge that is useful; hence, sometimes no decision is mandated.

Ronald P. Loui

For any system with limited statistical knowledge, the combination of evidence and the interpretation of sampling information require the determination of the right reference class (or of an adequate one). The present note (1) discusses the use of reference classes in evidential reasoning, and (2) discusses implementations of Kyburg's rules for reference classes. This paper contributes the first frank discussion of how much of Kyburg's system is needed to be powerful, how much can be computed effectively, and how much is philosophical fat.

Ronald P. Loui

This paper reports on empirical work aimed at comparing evidential reasoning techniques. While there is prima facie evidence for some conclusions, this i6 work in progress; the present focus is methodology, with the goal that subsequent results be meaningful. The domain is a network of UNIX* cycle servers, and the task is to predict properties of the state of the network from partial descriptions of the state. Actual data from the network are taken and used for blindfold testing in a betting game that allows abstention. The focal technique has been Kyburg's method for reasoning with data of varying relevance to a particular query, though the aim is to be able eventually to compare various uncertainty calculi. The conclusions are not novel, but are instructive. 1. All of the calculi performed better than human subjects, so unbiased access to sample experience is apparently of value. 2. Performance depends on metric: (a) when trials are repeated, net = gains - losses favors methods that place many bets, if the probability of placing a correct bet is sufficiently high; that is, it favors point-valued formalisms; (b) yield = gains/(gains + lossee) favors methods that bet only when sure to bet correctly; that is, it favors interval-valued formalisms. 3. Among the calculi, there were no clear winners or losers. Methods are identified for eliminating the bias of the net as a performance criterion and for separating the calculi effectively: in both cases by posting odds for the betting game in the appropriate way.

Ronald P. Loui

In two recent papers, I have proposed a description of decision analysis that differs from the Bayesian picture painted by Savage, Jeffrey and other classic authors. Response to this view has been either overly enthusiastic or unduly pessimistic. In this paper I try to place the idea in its proper place, which must be somewhere in between. Looking at decision analysis as defeasible reasoning produces a framework in which planning and decision theory can be integrated, but work on the details has barely begun. It also produces a framework in which the meta-decision regress can be stopped in a reasonable way, but it does not allow us to ignore meta-level decisions. The heuristics for producing arguments that I have presented are only supposed to be suggestive; but they are not open to the egregious errors about which some have worried. And though the idea is familiar to those who have studied heuristic search, it is somewhat richer because the control of dialectic is more interesting than the deepening of search.