A key concept in many Information Retrieval (IR) tasks, e.g. document indexing, query language modelling, aspect and diversity retrieval, is the relevance measurement of topics, i.e. to what extent an information object (e.g. a document or a query) is about the topics. This paper investigates the interference of relevance measurement of a topic caused by another topic. For example, consider that two user groups are required to judge whether a topic q is relevant to a document d, and q is presented together with another topic (referred to as a companion topic). If different companion topics are used for different groups, interestingly different relevance probabilities of q given d can be reached. In this paper, we present empirical results showing that the relevance of a topic to a document is greatly affected by the companion topic’s relevance to the same document, and the extent of the impact differs with respect to different companion topics. We further analyse the phenomenon from classical and quantum-like interference perspectives, and connect the phenomenon to nonreality and contextuality in quantum mechanics. We demonstrate that quantum like model fits in the empirical data, could be potentially used for predicting the relevance when interference exists.

The Predication-based Semantic Indexing (PSI) approach encodes both symbolic and distributional information into a semantic space using a permutation-based variant of Random Indexing. In this paper, we develop and evaluate a computational model of abductive reasoning based on PSI. Using distributional information, we identify pairs of concepts that are likely to be predicated about a common third concept, or middle term. As this occurs without the explicit identification of the middle term concerned, we refer to this process as a “logical leap”. Subsequently, we use further operations in the PSI space to retrieve this middle term and identify the predicate types involved. On evaluation using a set of 1000 randomly selected cue concepts, the model is shown to retrieve with accuracy concepts that can be connected to a cue concept by a middle term, as well as the middle term concerned, using nearest-neighbor search in the PSI space. The utility of quantum logical operators as a means to identify alternative paths through this space is also explored.

We assume that the brain is some kind of a computer and look at operations implied by the figurative use of language. Figurative language is pervasive, bypasses the literal meaning of what is said and is interpreted metaphorically or by analogy. Such an interpretation calls for a mapping in concept space, leading us to speculate about the nature of concept space in terms of readily computable mappings. We find that mappings of the appropriate kind are possible in high-dimensional spaces and demonstrate them with the simplest such space, namely, where the dimensions are binary. Two operations on binary vectors, one akin to addition and the other akin to multiplication, allow new representations to be composed from existing ones, and the ``multiplication'' operation is also suited for the mapping. The properties of high-dimensional spaces have been shown elsewhere to correspond to cognitive phenomena such as memory recall. The present ideas further suggest the suitability of high-dimensional representation for cognitive modeling.

Forming ad-hoc coalitions between military forces and humanitarian organizations is crucial in mission-critical scenarios. Very often coalition parties need to operate according to planning constraints and regulations, or policies. Therefore, they find themselves not only in need to consider their own goals, but also to support coalition partners to the extent allowed by such regulations. In time-stressed conditions, this is a challenging and cognition-intensive task. In this paper, we present intelligent agents that support human planners and ease their cognitive burden by detecting and giving advice about the violation of policies and constraints. Through a series of experiments conducted with human subjects, we compare and contrast the agents' performance on a number of metrics in three conditions: agent support, transparent policy enforcement, and neither support nor enforcement.

Journalists and editors work under pressure to collect relevant details and background information about specific events. They spend a significant amount of time sifting through documents and finding new information such as facts, opinions or stakeholders (i.e. people, places and organizations that have a stake in the news). Spotting them is a tedious and cognitively intense process. One task, essential to this process, is to find and keep track of stakeholders. This task is taxing cognitively and in terms of memory. Tell Me More offers an automatic aid to this task. Tell Me More is a system that, given a seed story, mines the web for similar stories reported by different sources and selects only those stories which offer new information with respect to that original seed story. Much like a journalist, the task of detecting named entities is central to its success. In this paper we briefly describe Tell Me More and, in particular, we focus on Tell Me More's entity detection component. We describe an approach that combines off-the-shelf named entity recognizers (NERs) with WPED, an in-house publicly available NER that uses Wikipedia as its knowledge base. We show significant increase in precision scores with respect to traditional NERs. Lastly, we present an overall evaluation of Tell Me More using this approach.

A promising application area for proactive assistant agents is automated tutoring and training.  Intelligent tutoring systems (ITSs) assist tutors and tutees by automating diagnosis and adaptive tutoring. These tasks are well modeled by a partially observable Markov decision process (POMDP) since it accounts for the uncertainty inherent in diagnosis. However, an important aspect of making POMDP solvers feasible for real-world problems is selecting appropriate representations for states, actions, and observations. This paper studies two scalable POMDP state and observation representations. State queues allow POMDPs to temporarily ignore less-relevant states. Observation chains represent information in independent dimensions using sequences of observations to reduce the size of the observation set. Preliminary experiments with simulated tutees suggest the experimental representations perform as well as lossless POMDPs, and can model much larger problems.

We consider an alignment algorithm for reconstructing global coordinates from local coordinates constructed for sections of manifolds. We show that, under certain conditions, the align- ment algorithm can successfully recover global coordinates even when local neighborhoods have different dimensions. Our results generalize an earlier analysis to allow alignment of sections of different dimensions. We also apply our result to a semisupervised learning problem.

The analysis of career paths suffers from a lack of exploratory tools and dynamic models, due in part to the inherent high dimensionality of the problem. Paths may be understood as directed traversals through a graph whose nodes consist of "job types," which we define as industry and occupation pairs. We want to develop tools to understand and detect high-level features of  both the labor market and the workers moving through it — career dynamics. To do this, we map the discrete space of jobs into a d-dimensional continuous space; proximity between jobs will mean that they are "close" to each other in a non-negligible subset of career paths. This embedding allows one to visualize the job landscape.  Moreover, we can map individual or groups of career paths to this space, extract features of their collective structure, and construct statistical tests comparing groups by means of this mapping.

The ability to take intelligent actions in real-world domains is a goal of great interest in the machine learning community. Unfortunately, the real-world is filled with systems that can bepartially observed but cannot, as yet, be described by first principlemodels. Moreover, the traditional paradigm of direct interaction with the environment used in reinforcement learning (RL) is often prohibitively expensive in practice. An alternative approach that simultaneously solves both of these problems is to gain experience in simulation; the simulation in this approach is a computational model derived from observations. Advances in sensory and information technology are simplifying the acquisition and distribution of real-world datasets to computational scientists; thus, the barrier to linking intelligent control with real-world domains is becoming one of identifying high-quality state-space and transition functions directly from observations. From a dynamical systems perspective, this barrier is analogous to the problem of finding high-quality manifold embeddings and a rich literature of theory and practice exists to address it. The contribution of this work is two-fold. First, we describe an approach for learning optimal control strategies directly from observations using manifold embeddings as the intermediate state representation. Second, we demonstrate how control strategies constructed in this way can answer important scientific questions. As a concrete example, we use our approach to guide experimental decisions in neurostimulation treatments of epilepsy.

Despite efforts to build robust vision systems, robots in new environments inevitably encounter new objects. Traditional supervised learning requires gathering and annotating sampleimages in the environment, usually in the form of bounding boxes or segmentations. This training interface takes some experience to do correctly and is quite tedious. We report work in progress on a robotic dialog system to learn names and attributes of objects through spoken interaction with a human teacher. The robot and human play a variant of the children’s games “I Spy” and “20 Questions”. In our game, the human places objects of interest in front of the robot, then picks an object in her head. The robot asks a series of natural language questions about the target object, with the goal of pointing at the correct object while asking a minimum number of questions. The questions range from attributes such as color (“Is it red?”) to category questions (“Is it a cup?”). The robot selects questions to ask based on an information gain criteria, seeking to minimize the entropy of the visual model given the answer to the question.