Many of the conditions in which technology is required to adapt cannot be anticipated during its design stage, creating a significant challenge for the designer. Inspired by the study of a range of biological systems, we propose that degeneracy - the realization of multiple, functionally versatile components with contextually overlapping functional redundancy - will support adaptation in technologies because it effects pervasive flexibility, evolutionary innovation, and homeostatic robustness. We provide examples of degeneracy in a number of rudimentary living technologies from military socio-technical systems to swarm robotics and we present design principles - including protocols, loose regulatory coupling, and functional versatility - that allow degeneracy to arise in both biological and man-made systems. Keywords: pervasive adaptation, degeneracy, living technologies, distributed robustness 1. Introduction Unanticipated requirements can arise throughout a technology's life and are a notoriously difficult engineering problem and a challenging research topic because past routines and contingency plans will be of limited utility. Dealing with new challenges requires exploration, diversity, and bethedging: principles that are common to any discipline in which responses to novelty determine competitive success.

We consider the problem of manipulating elections by cloning candidates. In our model, a manipulator can replace each candidate c by several clones, i.e., new candidates that are so similar to c that each voter simply replaces c in his vote with a block of these new candidates, ranked consecutively. The outcome of the resulting election may then depend on the number of clones as well as on how each voter orders the clones within the block. We formalize what it means for a cloning manipulation to be successful (which turns out to be a surprisingly delicate issue), and, for a number of common voting rules, characterize the preference profiles for which a successful cloning manipulation exists. We also consider the model where there is a cost associated with producing each clone, and study the complexity of finding a minimum-cost cloning manipulation. Finally, we compare cloning with two related problems: the problem of control by adding candidates and the problem of possible (co)winners when new alternatives can join.

We identify principles characterizing Solomonoff Induction by demands on an agent's external behaviour. Key concepts are rationality, computability, indifference and time consistency. Furthermore, we discuss extensions to the full AI case to derive AIXI.

This paper explores six different representations of the BlocksWorld Domain. It compares the results of seven planners run on these representations. It shows that the rankings for the International Planning Competition, using the non-satisficing scoring function, would change for every representation.

We propose a theory of abstraction of discrete-event systems (DES) formulated at the semantic level, i.e., as a function that maps event traces at the original (ground) level to traces at the abstract level. We study how diagnosis of DES can be performed using an abstract model, and under which conditions this process leads to a correct solution (i.e., a set of alternative diagnoses that include the real status of the system). Finally, we study how the use of an abstract model can affect the precision of diagnosis, i.e., the presence of spurious system states in the solution. To this end, we introduce the notion of diagnosability with abstract models, which ensures the precision of abstract diagnoses, and we discuss a practical way to test it.

Moreover, all of the of a system and, after detection, to determine the location faults that occurred within the (possibly extended) time interval and/or the type of system faults that caused the abnormal during which the system has been observed must be behaviour. A diagnosis hypothesis indicates which fault(s) accounted for in the diagnosis. Considering again the diagnosis occurred in the system, and the diagnosis is the set of alternative of a car, for each component we could be interested hypotheses that explain (i.e., are compatible) with in knowing whether a fault has occurred to it during the last the observed system behaviour. In this paper, we focus on week; in such a case, it is difficult to perform a drastic abstraction Model-Based Diagnosis (MBD) of Discrete-Event Systems of the model without losing any precision in the (DESs, see (Cassandras and Lafortune 1999)), where the diagnosis discrimination among different hypotheses. is computed by comparing a complete DES model In this article, we study a novel approach to reduce the of the system behaviour with a (partial) observation of the complexity of DES diagnosis, based on a reformulation of actual system behaviour (Sampath et al. 1995).

It is costly and time consuming to develop Multiple Choice Questions (MCQ) by hand. Using web-based resources to automate components of MCQ development would greatly benefit the education community through reducing reduplication of effort. Similar to many areas of Natural Language Processing (NLP), human-judged data is needed to train automated systems, but the majority of such data is proprietary. We present a graph-based representation for gathering training data from existing, web-based resources that increases access to such data and better directs the development of good questions.

Learning to use an object as a tool requires understanding what goals it helps to achieve, the properties of the tool that make it useful and how the tool must be manipulated to achieve the goal. We present a method that allows a robot to learn about objects in this way and thereby employ them as tools. An initial hypothesis for an action model of tool use is created by observing another agent accomplishing a task using a tool. The robot then refines its hypothesis by active learning, generating new experiments and observing the outcomes. Hypotheses are updated using Inductive Logic Programming. One of the novel aspects of this work is the method used to select experiments so that the search through the hypothesis space is minimised.

In this paper we address the design of robots that can be successful partners to humans in joint activity. The paper outlines an approach to achieving adjustable autonomy during execution- and hence to achieve resilient multi-actor joint action - based on both temporal and epistemic situation projection. The approach is based on non-deterministic planning techniques based on the situations calculus.

We take a conversational character trained on a set of linked question-answer pairs authored by hand, and augment its training data by adding sets of question-answer pairs which are generated automatically from texts on different topics. The augmented characters can answer questions about the new topics, at the cost of some performance loss on questions about the topics that the original character was trained to answer.