We consider the following sequential allocation process. A benevolent central authority has to allocate a set of indivisible goods to a set of agents whose preferences it is totally ignorant of. We consider the process of allocating objects one after the other by designating an agent and asking her to pick one of the objects among those that remain. The problem consists in choosing the "best" sequence of agents, according to some optimality criterion. We assume that agents have additive preferences over objects. The choice of an optimality criterion depends on three parameters: how utilities of objects are related to their ranking in an agent's preference relation; how the preferences of different agents are correlated; and how social welfare is defined from the agents' utilities. We address the computation of a sequence maximizing expected social welfare under several assumptions. We also address strategical issues.

In this paper, to support more precise Chinese Out-of-Vocabulary (OOV) term detection and Part-of-Speech (POS) guessing, a unified mechanism is proposed and formulated based on the fusion of multiple features and supervised learning. Besides all the traditional features, the new features for statistical information and global contexts are introduced, as well as some constraints and heuristic rules, which reveal the relationships among OOV term candidates. Our experiments on the Chinese corpora from both People’s Daily and SIGHAN 2005 have achieved the consistent results, which are better than those acquired by pure rule-based or statistics-based models. From the experimental results for combining our model with Chinese monolingual retrieval on the data sets of TREC-9, it is found that the obvious improvement for the retrieval performance can also be obtained.

Clause learning is a technique used by back-tracking-based propositional satisfiability solvers, where some clauses obtained by analysis of conflicts are added to the formula during backtracking. It has been observed empirically that clause learning does not significantly improve the performance of a solver when restricted to learning clauses of small width only. This experience is supported by lower bound theorems. It is shown that lower bounds on the runtime of width-restricted clause learning follow from lower bounds on the width of resolution proofs. This yields the first lower bounds on width-restricted clause learning for formulas in 3-CNF.

How do we scale information extraction to the massive size and unprecedented heterogeneity of the Web corpus? Beginning in 2003, our KnowItAll project has sought to extract high-quality knowledge from the Web. In 2007, we introduced the Open Information Extraction (Open IE) paradigm which eschews handlabeled training examples, and avoids domain-specific verbs and nouns, to develop unlexicalized, domain-independent extractors that scale to the Web corpus. Open IE systems have extracted billions of assertions as the basis for both common-sense knowledge and novel question-answering systems. This paper describes the second generation of Open IE systems, which rely on a novel model of how relations and their arguments are expressed in English sentences to double precision/recall compared with previous systems such as TEXTRUNNER and WOE.

Note that this extra to make decisions about which plans are used to information is included as a preference rather than a goal, achieve their goals. Usually the choice of which as it is acceptable to satisfy the goal without satisfying the plan to use to achieve a particular goal is left up preference. For example, if the user prefers to fly on Dodgy to the system to determine. In this paper we show Airlines, but no such flights are available, then specifying this how preferences, which can be set by the user of the as a preference means that the user can still have a holiday; system, can be incorporated into the BDI execution specifying this as a goal would mean that the user refuses to process and used to guide the choices made.

How obliged can we be to AI, and how much danger does it pose us? A surprising proportion of our society holds exaggerated fears or hopes for AI, such as the fear of robot world conquest, or the hope that AI will indefinitely perpetuate our culture. These misapprehensions are symptomatic of a larger problem—a confusion about the nature and origins of ethics and its role in society. While AI technologies do pose promises and threats, these are not qualitatively different from those posed by other artifacts of our culture which are largely ignored: from factories to advertising, weapons to political systems. Ethical systems are based on notions of identity, and the exaggerated hopes and fears of AI derive from our cultures having not yet accommodated the fact that language and reasoning are no longer uniquely human. The experience of AI may improve our ethical intuitions and self-understanding, potentially helping our societies make better-informed decisions on serious ethical dilemmas.

Designing dialog policies for voice-enabled interfaces is a tailoring job that is most often left to natural language processing experts. This job is generally redone for every new dialog task because cross-domain transfer is not possible. For this reason, machine learning methods for dialog policy optimization have been investigated during the last 15 years. Especially, reinforcement learning (RL) is now part of the state of the art in this domain. Standard RL methods require to test more or less random changes in the policy on users to assess them as improvements or degradations. This is called on policy learning. Nevertheless, it can result in system behaviors that are not acceptable by users. Learning algorithms should ideally infer an optimal strategy by observing interactions generated by a non-optimal but acceptable strategy, that is learning off-policy. In this contribution, a sample-efficient, online and off-policy reinforcement learning algorithm is proposed to learn an optimal policy from few hundreds of dialogues generated with a very simple handcrafted policy.

Qualitative reasoning about commonsense space often involves entities of different dimensions. We present a weak axiomatization of multidimensional qualitative space based on `relative dimension' and dimension-independent `containment' which suffice to define basic dimension-dependent mereotopological relations. We show the relationships to other meoreotopologies and to incidence geometry. The extension with betweenness, a primitive of relative position, results in a first-order theory that qualitatively abstracts ordered incidence geometry.

One approach for artificially intelligent agents wishing to maximise some performance metric in a given domain is to learn from a collection of training data that consists of actions or decisions made by some expert, in an attempt to imitate that expert's style. We refer to this type of agent as an expert imitator. In this paper we investigate whether performance can be improved by combining decisions from multiple expert imitators. In particular, we investigate two existing approaches for combining decisions. The first approach combines decisions by employing ensemble voting between multiple expert imitators. The second approach dynamically selects the best imitator to use at runtime given the performance of the imitators in the current environment. We investigate these approaches in the domain of computer poker. In particular, we create expert imitators for limit and no limit Texas Hold'em and determine whether their performance can be improved by combining their decisions using the two approaches listed above.

Monte Carlo tree search (MCTS) methods have had recent success in games, planning, and optimization. MCTS uses results from rollouts to guide search; a rollout is a path that descends the tree with a randomized decision at each ply until reaching a leaf. MCTS results can be strongly influenced by the choice of appropriate policy to bias the rollouts. Most previous work on MCTS uses static uniform random or domain-specific policies. We describe a new MCTS method that dynamically adapts the rollout policy during search, in deterministic optimization problems. Our starting point is Cazenave's original Nested Monte Carlo Search (NMCS), but rather than navigating the tree directly we instead use gradient ascent on the rollout policy at each level of the nested search. We benchmark this new Nested Rollout Policy Adaptation (NRPA) algorithm and examine its behavior. Our test problems are instances of Crossword Puzzle Construction and Morpion Solitaire. Over moderate time scales NRPA can substantially improve search efficiency compared to NMCS, and over longer time scales NRPA improves upon all previous published solutions for the test problems. Results include a new Morpion Solitaire solution that improves upon the previous human-generated record that had stood for over 30 years.