We present the first sample compression algorithm for nearest neighbors with non-trivial performance guarantees. We complement these guarantees by demonstrating almost matching hardness lower bounds, which show that our bound is nearly optimal. Our result yields new insight into margin-based nearest neighbor classification in metric spaces and allows us to significantly sharpen and simplify existing bounds. Some encouraging empirical results are also presented.

Given an $n$-sample of random vectors $(X_i,Y_i)_{1 \leq i \leq n}$ whose joint law is unknown, the long-standing problem of supervised classification aims to \textit{optimally} predict the label $Y$ of a given a new observation $X$. In this context, the nearest neighbor rule is a popular flexible and intuitive method in non-parametric situations. Even if this algorithm is commonly used in the machine learning and statistics communities, less is known about its prediction ability in general finite dimensional spaces, especially when the support of the density of the observations is $\mathbb{R}^d$. This paper is devoted to the study of the statistical properties of the nearest neighbor rule in various situations. In particular, attention is paid to the marginal law of $X$, as well as the smoothness and margin properties of the \textit{regression function} $\eta(X) = \mathbb{E}[Y | X]$. We identify two necessary and sufficient conditions to obtain uniform consistency rates of classification and to derive sharp estimates in the case of the nearest neighbor rule. Some numerical experiments are proposed at the end of the paper to help illustrate the discussion.

We present a case-based approach to character identification in natural language text in the context of our Voz system. Voz first extracts entities from the text, and for each one of them, computes a feature-vector using both linguistic information and external knowledge. We propose a new similarity measure called Continuous Jaccard that exploits those feature-vectors to compute the similarity between a given entity and those in the case-base, and thus determine which entities are characters or not. We evaluate our approach by comparing it with different similarity measures and feature sets. Results show an identification accuracy of up to 93.49%, significantly higher than recent related work.

We address two domains of skill transfer problems encountered by an autonomous robot: within-domain adaptation and cross-domain transfer. Our aim is to provide skill representations which enable transfer in each problem classification. As such, we explore two approaches to skill representation which address each problem classification separately. The first representation, based on mimicking, encodes the full demonstration and is well suited for within-domain adaptation. The second representation is based on imitation and serves to encode a set of key points along the trajectory, which represent the goal points most relevant to the successful completion of the skill. This representation enables both within-domain and cross-domain transfer. A planner is then applied to these constraints, generating a domain-specific trajectory which addresses the transfer task.

Multi-Output Dependence (MOD) learning is a generalization of standard classification problems that allows for multiple outputs that are dependent on each other. A primary issue that arises in the context of MOD learning is that for any given input pattern there can be multiple correct output patterns. This changes the learning task from function approximation to relation approximation. Previous algorithms do not consider this problem, and thus cannot be readily applied to MOD problems. To perform MOD learning, we introduce the Hierarchical Multi-Output Nearest Neighbor model (HMONN) that employs a basic learning model for each output and a modified nearest neighbor approach to refine the initial results.

A Survey of Artificial Intelligence Research at the IIIA

A Survey of Artificial Intelligence Research at the IIIA Abstract The IIIA is a public research centre, belonging to the Spanish National Research Council (CSIC), dedicated to AI research. We focus our activities on a few well-defined sub-domains of Artificial Intelligence, positively avoiding dispersion and keeping a good balance between basic research and applications, and paying particular attention to training PhD students and technology transfer. In this article, we survey some of the most relevant results we have obtained during the last 12 years.

This page includes forthcoming AAAI sponsored conferences, conferences presented by AAAI Affiliates, and conferences held in cooperation with AAAI. AI Magazine also maintains a calendar listing that includes nonaffiliated conferences at www.aaai.org/Magazine/calendar.php. AIIDE-14 will be held FLAIRS-15 will be held May 18-20, 10th ACM/IEEE International Conference October 3-7 in Raleigh, NC, USA 2015 in Hollywood, Florida, USA on Human-Robot Interaction. ICAART 2014 will be held January 10-12 in Lisbon, Portugal International Joint Conference on AAAI Fall Symposium Series. ICCBR 2014 held January 10-12 in Lisbon, Portugal will be held September 29 - October 1 AAAI Spring Symposium.

It was founded in 1991 and, since 1994, has been located on the campus of the Autonomous University of Barcelona. IIIA grew out of an AI research group at the Center for Advanced Studies in Blanes (Spain) that started AI research in 1985. On average IIIA has had about 50 members per year during the last 12 years with a peak of almost 80 members in 2012. In total around 200 different people, including visiting researchers as well as master's and Ph.D. students, have been members of IIIA over the past 20 years. Seventy-seven students have completed their Ph.D. work at our Institute, 48 of them during the last 12 years.

Robots are added to human teams to increase the team's skills or capabilities but in order to get the full benefit the teams must trust the robots. We present an approach that allows a robot to estimate its trustworthiness and adapt its behavior accordingly. Additionally, the robot uses case-based reasoning to store previous behavior adaptations and uses this information to perform future adaptations. In a simulated robotics domain, we compare case-based behavior adaption to behavior adaptation that does not learn and show it significantly reduces the number of behaviors that need to be evaluated before a trustworthy behavior is found.