We study several classes of interactive assistants from the points of view of decision theory and computational complexity. We first introduce a special class of POMDPs called hidden-goal MDPs (HGMDPs), which formalize the problem of interactively assisting an agent whose goal is hidden and whose actions are observable. In spite of its restricted nature, we show that optimal action selection in finite horizon HGMDPs is PSPACE-complete even in domains with deterministic dynamics. We then introduce a more restricted model called helper action MDPs (HAMDPs), where the assistant’s action is accepted by the agent when it is helpful, and can be easily ignored by the agent otherwise. We show classes of HAMDPs that are complete for PSPACE and NP along with a polynomial time class. Furthermore, we show that for general HAMDPs a simple myopic policy achieves a regret, compared to an omniscient assistant, that is bounded by the entropy of the initial goal distribution. A variation of this policy is also shown to achieve worst-case regret that is logarithmic in the number of goals for any goal distribution.

This paper examines the problem of finding articles in Wikipedia to match noun synsets in WordNet. The motivation is that these articles enrich the synsets with much more information than is already present in WordNet. Two methods are used. The first is title matching, following redirects and disambiguation links. The second is information retrieval over the set of articles. The methods are evaluated over a random sample set of 200 noun synsets which were manually annotated. With 10 candidate articles retrieved for each noun synset, the methods achieve recall of 93%. The manually annotated data set and the automatically generated candidate article sets are available online for research purposes.

We have been exploring the use of Web-derived knowledge bases through the development of Wikitology — a hybrid knowledge base of structured and unstructured information extracted from Wikipedia augmented by RDF data from DBpedia and other Linked Open Data resources. In this paper, we describe approaches that aid in enriching Wikipedia and thus the resources that derive from Wikipedia such as the Wikitology knowledge base, DBpedia, Freebase and Powerset.

Present-day programs are brittle as computers are notoriously lacking in common sense. While significant progress has been made in building large common sense knowledge bases, they are intrinsically incomplete and inconsistent. This paper presents a novel approach to bridging the gaps between multiple knowledge bases, making it possible to answer queries based on knowledge collected from multiple sources without a common ontology. New assertions are found by computing graph similarity with principle component analysis to draw analogies across multiple knowledge bases. Experiments are designed to find new assertions for a Chinese commonsense knowledge base using the OMCS ConceptNet and similarly for WordNet. The assertions are voted by online users to verify that 75.77% / 77.59% for Chinese ConceptNet / WordNet respectively are good, despite the low overlap in coverage among the knowledge bases.

Increasingly, we need to computationally understand real-time streams of information in places such as news feeds, speech streams, and social networks. We present Streaming AnalogySpace, an efficient technique that discovers correlations in and makes predictions about sparse natural-language data that arrives in a real-time stream. AnalogySpace is a noise-resistant PCA-based inference technique designed for use with collaboratively collected common sense knowledge and semantic networks. Streaming AnalogySpace advances this work by computing it incrementally using CCIPCA, and keeping a dense cache of recently-used features to efficiently represent a sparse and open domain. We show that Streaming AnalogySpace converges to the results of standard AnalogySpace, and verify this by evaluating its accuracy empirically on common-sense predictions against standard AnalogySpace.

It has generally been assumed that the knowledge employed by an AI reasoning system needs to be in an unambiguous, formally interpretable form. From that perspective, general knowledge expressed in ordinary language (e.g., “dogs bark”) is unacceptably ambiguous and incomplete. However, we can achieve at least a partial transformation of such knowledge into formal, generically quantified sentences by taking account of properties of words and phrases such as the aspectual category, tense, Levin class, and presuppositions of verbs, or the classification of predicates (adjectival, nominal, verbal) as applicable to objects or kinds of objects.

Until recently, the AI and in particular the NLP community GA, immediately preceding the Twenty-Fourth AAAI Conference have relied on resources built manually by experts in on Artificial Intelligence -- AAAI 2010. It is a successor specific areas (in particular linguists, philosophers, cognitive to the workshops organized at AAAI 2008 entitled linguists). User contributed knowledge has opened up "Wikipedia and Artificial Intelligence: An Evolving Synergy" a new perspective, in that it captures the kind of knowledge (WikiAI 08) and at IJCAI 2009 entitled "User Contributed and organization that arises naturally out of the consensus Knowledge and Artificial Intelligence: An Evolving of the masses, and as such represents better our collective Synergy" (WikiAI 09). The outcome is a multifaceted and extremely This volume contains papers accepted for presentation at rich source of information, revealed through embedded annotations the workshop. We issued calls for regular papers, short latebreaking and structural information.

Dynamic executio n is a flexible plan execution technique in which a plan executive schedules and executes tasks dynamically at runtime in response to disturbances in order to satisfy plan constraints. In this paper, we extend dynamic execution to temporally and spatially flexible plans which, 1) execute tasks conditionally based on runtime state, and 2) support error recovery for anticipated runtime constraint violations. To accomplish these goals, we broaden our focus from dynamic execution of flexible plans to dynamic execution of flexible reactive programs. First, we introduce the Reactive Model-based Programming Language (RMPL) which, in addition to modeling temporal and spatial flexibility, includes three reactive programming language constructs: conditional execution, iteration, and exception handling. Then, we develop a probabilistic particle-sampling based dynamic execution algorithm which reasons efficiently over future program states to schedule tasks dynamically at runtime in order to satisfy program constraints. In addition, the algorithm monitors its own progress and notifies the executive if at any time the likelihood of successful program execution drops below a specified probability bound, δ.

In this paper we outline an approach to the integration of task planning and motion planning that has the following key properties: It is aggressively hierarchical. It makes choices and commits to them in a top-down fashion in an attempt to limit the length of plans that need to be constructed, and thereby exponentially decrease the amount of search required. Importantly, our approach also limits the need to project the effect of actions into the far future. It operates on detailed, continuous geometric representations and partial symbolic descriptions. It does not require a complete symbolic representation of the input geometry or of the geometric effect of the task-level operations.

Solving real-world problems using symbolic planning often requires a simplified formulation of the original problem, since certain subproblems cannot be represented at all or only in a way leading to inefficiency. For example, manipulation planning may appear as a subproblem in a robotic planning context or a packing problem can be part of a logistics task. In this paper we propose an extension of PDDL for specifying semantic attachments. This allows the evaluation of grounded predicates, the change of fluents and the calculation of durations by externally specified functions. Furthermore, we describe a general schema of integrating semantic attachments into forward-chaining planning systems and report on our experience of adding this extension to the planner Temporal Fast Downward. Finally, we present some preliminary experiments using semantic attachments.