We have developed a simulation model that accepts instructions in unconstrained natural language, and then guides a robot to the correct destination. The instructions are segmented on the basis of the actions to be taken, and each segment is labeled with the required action. This flat formulation reduces the problem to a sequential labeling task, to which machine learning methods are applied. We propose an innovativemachine learningmethod for explicitly modeling the actions described in instructions and integrating learning and inference about the physical environment. We obtained a corpus of 840 route instructions that experimenters verified as follow-able, given by people in building navigation situations. Using the four-fold cross validation, our experiments showed that the simulated robot reached the correct destination 88% of the time.

Particle physics experiments, like the Large Hadron Collider in Geneva, can generate thousands of data points listing detected particle reactions. An important learning task is to analyze the reaction data for evidence of conserved quantities and hidden particles. This task involves latent structure in two ways: first, hypothesizing hidden quantities whose conservation determines which reactions occur, and second, hypothesizing the presence of hidden particles. We model this problem in the classic linear algebra framework of automated scientific discovery due to Valdes-Perez, Zytkow and Simon, where both reaction data and conservation laws are represented as matrices. We introduce a new criterion for selecting a matrix model for reaction data: find hidden particles and conserved quantities that rule out as many interactions among the nonhidden particles as possible. A polynomial-time algorithm for optimizing this criterion is based on the new theorem that hidden particles are required if and only if the Smith Normal Form of the reaction matrix R contains entries other than 0 or 1. To our knowledge this is the first application of Smith matrix decomposition to a problem in AI. Using data from particle accelerators, we compare our algorithm to the main model of particles in physics, known as the Standard Model: our algorithm discovers conservation laws that are equivalent to those in the Standard Model, and indicates the presence of a  hidden particle (the electron antineutrino) in accordance with the Standard Model.

A method for expressive melody synthesis is presented seeking to capture the prosodic (stress and directional) element of musical interpretation. An expressive performance is represented as a note-level annotation, classifying each note according to a small alphabet of symbols describing the role of the note within a larger context.  An audio performance of the melody is represented in terms of two time-varying functions describing the evolving frequency and intensity.  A method is presented that transforms the expressive annotation into the frequency and intensity functions, thus giving the audio performance. The problem of expressive rendering is then cast as estimation of the most likely sequence of hidden variables corresponding to the prosodic annotation. Examples are presented on a dataset of around 50 folk-like melodies, realized both from hand-marked and estimated annotations.

This paper presents a novel approach to the estimation of user's affective states in Human-Computer Interaction. Most of the present approaches divide emotions strictly between positive or negative. However, recent discoveries in the field of Emotional Intelligence show that emotions should be rather perceived as context-sensitive engagements with the world. This leads to a need to specify whether the emotions conveyed in a conversation are appropriate for a situation they are expressed in. In the proposed method we use a system for affect analysis on textual input to recognize users’ emotions and a Web mining technique to verify the contextual appropriateness of those emotions. On this basis a conversational agent can choose to either sympathize with the user or help them manage their emotions. Finally, the results of evaluation of the proposed method with two different conversational agents are discussed, and perspectives for further development of the method are proposed.

There is increasing interest in building systems that can automatically interpret hand-drawn sketches. However, many challenges remain in terms of recognition accuracy, robustness to different drawing styles, and ability to generalize across multiple domains. To address these challenges, we propose a new approach to sketched symbol recognition that focuses on the visual appearance of the symbols. This allows us to better handle the range of visual and stroke-level variations found in freehand drawings. We also present a new symbol classifier that is computationally efficient and invariant to rotation and local deformations. We show that our method exceeds state-of-the-art performance on all three domains we evaluated, including handwritten digits, PowerPoint shapes, and electrical circuit symbols.

We investigate prediction of users' desktop activities in the Unix domain. The learning techniques we explore do not require explicit user teaching. We show that simple efficient many-class learning can perform well for action prediction, significantly improving over previously published results and baselines. This finding is promising for various human-computer interaction scenarios where a rich set of potentially predictive features is available, where there can be many different actions to predict, and where there can be considerable nonstationarity.

As data-center energy consumption continues to rise, efficient power management is becoming increasingly important. In this work, we examine the use of a novel market mechanism for finding the right balance between power and performance. The market enables a separation between a `buyer side' that strives to maximize performance and a 'seller side' that strives to minimize power and other costs. A concise and scalable description language is defined for agent preferences that admits a mixed-integer program for computing optimal allocations. Experimental results demonstrate the robustness, flexibility, practicality and scalability of the architecture.

The Berkeley Drosophila Genome Project (BDGP) has produced a large number of gene expression patterns, many of which have been annotated textually with anatomical and developmental terms. These terms spatially correspond to local regions of the images; however, they are attached collectively to groups of images, such that it is unknown which term is assigned to which region of which image in the group. This poses a challenge to the development of the computational method to automate the textual description of expression patterns contained in each image. In this paper, we show that the underlying nature of this task matches well with Figure 1: Samples of images and associated annotation terms a new machine learning framework, Multi-Instance of the gene Actn in the stage ranges 11-12 and 13-16 in the Multi-Label learning (MIML). We propose a new BDGP database. The darkly stained region highlights the MIML support vector machine to solve the problems place where the gene is expressed. The darker the region, that beset the annotation task.

This paper deals with the relationship between intelligent behaviour, on the   one hand, and the mental qualities needed to produce it, on the other.  We   consider two well-known opposing positions on this issue: one due to Alan   Turing and one due to John Searle (via the Chinese Room).  In particular, we   argue against Searle, showing that his answer to the so-called System Reply   does not work.  The argument takes a novel form:   we shift the debate to a different and more plausible room where the   required conversational behaviour is much easier to characterize and to   analyze.  Despite being much simpler than the Chinese Room, we show that    the  behaviour there is still complex enough that it cannot be produced without   appropriate mental qualities.

Written instructions are a common way of teaching people how to accomplish tasks on the web. However, studies have shown that written instructions are difficult to follow, even for experienced users. A system that understands human-written instructions could guide users through the process of following the directions, improving completion rates and enhancing the user experience. While general natural language understanding is extremely difficult, we believe that in the limited domain of how-to instructions it should be possible to understand enough to provide guided help in a mixed-initiative environment. Based on a qualitative analysis of instructions gathered for 43 web-based tasks, we have formalized the problem of understanding and interpreting how-to instructions. We compare three different approaches to interpreting instructions: a keyword-based interpreter, a grammar-based interpreter, and an interpreter based on machine learning and information extraction. Our empirical results demonstrate the feasibility of automated how-to instruction understanding.