John Pollock

Decision-theoretic planning is normally based on the assumption that plans can be compared by comparing their expected-values, and the objective is to find an optimal plan. This is typically defended by reference to classical decision theory. However, classical decision theory is actually incompatible with this “simple plan-based decision theory”. A defense of plan-based decision theory must begin by showing that classical decision theory is incorrect insofar as the two theories conflict, so this paper begins by raising objections to classical decision theory. First, there is a discussion of the considerations arising out of the Newcomb problem that have given rise to causal decision theory. Next, counterexamples are constructed for classical decision theory turning on the fact that an agent may be unable to perform an action, and may even be unable to try to perform an action. A proposal is made for how to repair classical decision theory in light of these counterexamples. But then turning to the concept of an “alternative” that is presupposed by classical decision theory, it is argued that actions must often be chosen in groups rather than individually, i.e., the objects of rational choice are plans. It is argued that optimality cannot be defined for plans, and even if it could be, it would not be reasonable to require rational agents to find optimal plans. So simple plan-based decision theory must also be rejected. An alternative called “locally global planning” is proposed as a replacement for both classical decision theory and simple plan-based decision theory

There are two general approaches to handling contingencies in decision-theoretic planning. State-space planners reason globally, building a map of the parts of the world relevant to the planning problem, and then attempt to distill a plan out of the map. POCL planners reason locally, attempting to build the plan up from local relationships. A planning problem is constructed that humans find trivial, but no state-space planner can solve. This motivates an investigation of decision-theoretic POCL contingency planners. Existing POCL contingency planners attempt to generalize the results of classical POCL contingency planning. However, this paper argues that the nature of contingency planning changes dramatically in decision-theoretic contexts, and results from classical contingency planning are of little relevance. In particular, in classical planning contingencies can only be attached to conditional forks, but in most uses of contingencies in decision-theoretic planning they are attached to single branches of the plan rather than to conditional forks. A criterion of adequacy for contingency planners is formulated, following from ordinary completeness, and it is shown that existing decision-theoretic POCL contingency planners do not satisfy it. Some tentative suggestions are made regarding how to construct a planner that does satisfy the adequacy condition.

The objective of this paper is to construct an implementable theory of rational decision-making for cognitive agents subject to realistic resource constraints. It is argued that decision-making should select actions indirectly by selecting plans that prescribe them. It is also argued that although expected values provide the tool for evaluating plans, plans cannot be compared straightforwardly in terms of their expected values, and the objective of a realistic agent cannot be to find optimal plans. The theory of Locally Global planning is proposed as a realistic alternative to standard "maximizing" theories of rational decision-making.

In concrete applications of probability, statistical investigation gives us knowledge of some probabilities, but we generally want to know many others that are not directly revealed by our data. For instance, we may know prob(P/Q) (the probability of P given Q) and prob(P/R), but what we really want is prob(P/Q&R), and we may not have the data required to assess that directly. The probability calculus is of no help here. Given prob(P/Q) and prob(P/R), it is consistent with the probability calculus for prob(P/Q&R) to have any value between 0 and 1. Is there any way to make a reasonable estimate of the value of prob(P/Q&R)? A related problem occurs when probability practitioners adopt undefended assumptions of statistical independence simply on the basis of not seeing any connection between two propositions. This is common practice, but its justification has eluded probability theorists, and researchers are typically apologetic about making such assumptions. Is there any way to defend the practice? This paper shows that on a certain conception of probability — nomic probability — there are principles of “probable probabilities” that license inferences of the above sort. These are principles telling us that although certain inferences from probabilities to probabilities are not deductively valid, nevertheless the second-order probability of their yielding correct results is 1. This makes it defeasibly reasonable to make the inferences. Thus I argue that it is defeasibly reasonable to assume statistical independence when we have no information to the contrary. And I show that there is a function Y(r,s:a) such that if prob(P/Q) = r, prob(P/R) = s, and prob(P/U) = a (where U is our background knowledge) then it is defeasibly reasonable to expect that prob(P/Q&R) = Y(r,s:a). Numerous other defeasible inferences are licensed by similar principles of probable probabilities..

Since Gettier, much of epistemology has focused on analyzing “S knows that P”, but that is not my interest. My general interest is in rational cognition — both in what it is to be rational, and in how rational cognition works. The traditional epistemological question, “How do you know?”, can be taken as addressing part of the more general problem of producing a theory of rational cognition. It is about specifically epistemic rationality. I interpret this question literally, as a question about how we should proceed in our epistemic endeavors. Epistemological theories that try to answer this question are theories of procedural epistemology (see my 1998), and when, from this perspective, we assess beliefs in terms of their justifiedness, the concept of justification is one of procedural epistemic justification. Whether this has anything to do with the analysis of knowledge is an open question, and not one that I have much interest in addressing.

When combining information from multiple sources and attempting to estimate the probability of a conclusion, we often find ourselves in the position of knowing the probability of the conclusion conditional on each of the individual sources, but we have no direct information about the probability of the conclusion conditional on the combination of sources. The probability calculus provides no way of computing such joint probabilities. This paper introduces a new way of combining probabilistic information to estimate joint probabilities. It is shown that on a particular conception of objective probabilities, clear sense can be made of second-order probabilities (probabilities of probabilities), and these can be related to combinatorial theorems about proportions in finite sets as the sizes of the sets go to infinity. There is a rich mathematical theory consisting of such theorems, and the theorems generate corresponding theorems about secondorder probabilities. Among the latter are a number of theorems to the effect that certain inferences from probabilities to probabilities, although not licensed by the probability calculus, have probability 1 of producing correct results. This does not mean that they will always produce correct results, but the set of cases in which the inferences go wrong form a set of measure 0. Among these inferences are some enabling us to reasonably estimate the values of joint probabilities in a wide range of cases. A function called the Y-function is defined. The central theorem is the Y-Theorem, which tells us that if we know the individual probabilities for the different information sources and estimate the joint probability using the Y-function, the second-order probability of getting the right answer is 1. This mathematical result is tested empirically using a simple multi-sensor example. The Y-theorem agrees with Dempster's rule of combination in special cases, but not in general.

I argue here that sophisticated AI systems, with the exception of those aimed at the psychological modeling of human cognition, must be based on general philosophical theories of rationality and, conversely, philosophical theories of rationality should be tested by implementing them in AI systems. So the philosophy and the AI go hand in hand. I compare human and generic rationality within a broad philosophy of AI and conclude by suggesting that ultimately, virtually all familiar philosophical problems will turn out to be at least indirectly relevant to the task of building an autonomous rational agent, and conversely, the AI enterprise has the potential to throw light at least indirectly on most philosophical problems.

Stuart Russell [14] describes rational agents as --œthose that do the right thing--�. The problem of designing a rational agent then becomes the problem of figuring out what the right thing is. There are two approaches to the latter problem, depending upon the kind of agent we want to build. On the one hand, anthropomorphic agents are those that can help human beings rather directly in their intellectual endeavors. These endeavors consist of decision making and data processing. An agent that can help humans in these enterprises must make decisions and draw conclusions that are rational by human standards of rationality. Anthropomorphic agents can be contrasted with goal-oriented agents --” those that can carry out certain narrowly-defined tasks in the world. Here the objective is to get the job done, and it makes little difference how the agent achieves its design goal.

Imagine yourself sitting on your front porch, sipping your morning coffee and admiring the scene before you. You see trees, houses, people, automobiles; you see a cat running across the road, and a bee buzzing among the flowers. You see that the flowers are yellow, and blowing in the wind. You see that the people are moving about, many of them on bicycles. You see that the houses are painted different colors, mostly earth tones, and most are one-story but a few are two-story. It is a beautiful morning. Thus the world interfaces with your mind through your senses. There is a strong intuition that we are not disconnected from the world. We and the other things we see around us are part of a continuous whole, and we have direct access to them through vision, touch, etc. However, the philosophical tradition tries to drive a wedge between us and the world by insisting that the information we get from perception is the result of inference from indirect evidence that is about how things look and feel to us. The philosophical problem of perception is then to explain what justifies these inferences. We will focus on visual perception. Figure one presents a crude diagram of the cognitive system of an agent capable of forming beliefs on the basis of visual perception. Cognition begins with the stimulation of the rods and cones on the retina. From that physical input, some kind of visual processing produces an introspectible visual image. In response to the production of the visual image, the cognizer forms beliefs about his or her surroundings. Some beliefs the perceptual beliefs are formed as direct responses to the visual input, and other beliefs are inferred from the perceptual beliefs. The perceptual beliefs are, at the very least, caused or causally influenced by having the image. This is signified by the dashed arrow marked with a large question mark. We will refer to this as the mystery link. Figure one makes it apparent that in order to fully understand how knowledge is based on perception, we need three different theories..

Human beings think of themselves in terms of a privileged non-descriptive designator — a mental “I”. Such thoughts are called “_de se_” thoughts. The mind/body problem is the problem of deciding what kind of thing I am, and it can be regarded as arising from the fact that we think of ourselves non-descriptively. Why do we think of ourselves in this way? We investigate the functional role of “I” (and also “here” and “now”) in cognition, arguing that the use of such non-descriptive “reflexive” designators is essential for making sophisticated cognition work in a general-purpose cognitive agent. If we were to build a robot capable of similar cognitive tasks as humans, it would have to be equipped with such designators

The contributions in this volume make an important effort to resurrect a rather old fashioned form of foundationalism. They defend the position that there are some beliefs that are justified, and are not themselves justified by any further beliefs. This epistemic foundationalism has been the subject of rigorous attack by a wide range of theorists in recent years, leading to the impression that foundationalism is a thing of the past. DePaul argues that it is precisely the volume and virulence of the assaults which points directly to the strength and coherence of the position

The strategy of this paper is to throw light on rational cognition and epistemic justification by examining irrationality. Epistemic irrationality is possible because we are reflexive cognizers, able to reason about and redirect some aspects of our own cognition. One consequence of this is that one cannot give a theory of epistemic rationality or epistemic justification without simultaneously giving a theory of practical rationality. A further consequence is that practical irrationality can affect our epistemic cognition. I argue that practical irrationality derives from a general difficulty we have in overriding built-in shortcut modules aimed at making cognition more efficient, and all epistemic irrationality can be traced to this same source. A consequence of this account is that a theory of rationality is a descriptive theory, describing contingent features of a cognitive architecture, and it forms the core of a general theory of “voluntary” cognition — those aspects of cognition that are under voluntary control. It also follows that most of the so-called “rules for rationality” that philosophers have proposed are really just rules describing default (non- reflexive) cognition. It can be perfectly rational for a reflexive cognizer to break these rules. The “normativity” of rationality is a reflection of a built-in feature of reflexive cognition — when we detect violations of rationality, we have a tendency to desire to correct them. This is just another part of the descriptive theory of rationality. Although theories of rationality are descriptive, the structure of reflexive cognition gives philosophers, as human cognizers, privileged access to certain aspects of rational cognition. Philosophical theories of rationality are really scientific theories, based on inference to the best explanation, that take contingent introspective data as the evidence to be explained.

Postulational approaches attempt to understand the dynamics of belief revision by appealing to no more than the set of beliefs held by an agent and the logical relations between them. It is argued there that such an approach cannot work. A proper account of belief revision must also appeal to the arguments supporting beliefs, and recognize that those arguments can be defeasible. If we begin with a mature epistemological theory that accommodates this, it can be seen that the belief revision operators on which the postulational theories are based are ill-defined. It is further argued that there is no way to repair the definitions so as to retain the spirit of those theory. Belief revision is better studied from within an independently motivated epistemological theory.

Probabilities are important in belief updating, but probabilistic reasoning does not subsume everything else (as the Bayesian would have it). On the contrary, Bayesian reasoning presupposes knowledge that cannot itself be obtained by Bayesian reasoning, making generic Bayesianism an incoherent theory of belief updating. Instead, it is indefinite probabilities that are of principal importance in belief updating. Knowledge of such indefinite probabilities is obtained by some form of statistical induction, and inferences to non-probabilistic conclusions are carried out in accordance with the statistical syllogism. Such inferences have been the focus of much attention in the nonmonotonic reasoning literature, but the logical complexity of such inference has not been adequately appreciated