This book collects all the papers that appeared in 2005 in Kno- edge, Rationality and Action (KRA), a journal published as a s- cial section of Synthese, which addresses contemporary issues in epistemic logic, belief revision, game and decision theory, rational agency, planning and theories of action. As such, the special section appeals to researchers from Computer Science, Game Theory, Ar- ?cial Intelligence, Philosophy, Knowledge Representation, Logic and Agents, addressing issues in arti?cial systems that have to gather information, reason about it and then make a sensible decision about what to do next. It will be clear already from the contents pages, that this book indeed re?ects the core of KRA: the papers in this volume address degrees of belief or certainty, and rational agency. The latter has several manifestations: often constraints on the agent’s belief, behaviour or decision making. Moreover, this book shows that KRA indeed represents a ‘loop’ in the behaviour of the agent: after having made a decision, the life of the agent does not end, rather, it will do some sensing or collect otherwise the outcome of its decision, to update its beliefs or knowledge accordingly and make up its mind about the next decision task.
This book is about Rational Agents, which can be humans, players in a game, software programs or institutions. Typically, such agents are uncertain about the state of affairs or the state of other agents, and under this partial information they have to decide on which action to take next.
This book collects chapters that give formal accounts not only of Uncertainty, Rationality and Agency, but also of their interaction: what are rational criteria to accept certain beliefs, or to modify them; how can degrees of beliefs guide an agent in making decisions; why distinguish between practical and epistemic rationality when agents try to coordinate; what must be common beliefs between agents about each other's rationality in order to act rationally themselves; can an agent assign probabilities to planned actions; how to formalise assumptions about a rational speaker in a conversation obeying Gricean maxims; how should a rational agent best represent the states, consequences, and acts that constitute the agent's rational decision problem?
This volume should appeal to researchers addressing issues in artificial systems that have to gather information in order to obtain Knowledge, reason about it and then make a Rational decision about which Action to take next.
