Modeling interaction with robots raises new and different challenges for dialog modeling than traditional dialog modeling with less embodied machines. We present four case studies of implementing a typical human-robot interaction scenario with different state-of-the-art dialog frameworks in order to identify challenges and pitfalls specific to HRI and potential solutions. The results are discussed with a special focus on the interplay between dialog and task modeling on robots.

In parallel with these efforts, significant advances have also been made in robotics. Innovations in sensing, reasoning, and manipulation have allowed autonomous robots to move beyond the walls of computing labs into the workplace, home, and street. Bringing robots into real-world environments has made it clear to researchers that robots need not only accurately navigate and manipulate objects, but also to work alongside and, ultimately, interact and collaborate with humans. Subsequently, efforts at the intersection of spoken dialogue and human-robot interaction (HRI) have sought to broaden studies of spoken dialogue to richer, more natural, physically situated settings, and have brought to the fore the rich research area of situated dialogue, focused on challenges and opportunities at the intersection of natural language, robotics, and commonsense reasoning. Projects in this realm have addressed challenges with the use of dialogue as enabling coordination among multiple actors, taking into consideration not only the details of the task at hand, but also the dynamic physical and social context in which the actors are immersed and the affordances that embodiment provides. This special issue of AI Magazine on dialogue with robots brings together a collection of articles on situated dialogue.

Modeling interaction with robots raises new and different challenges for dialog modeling than traditional dialog modeling with less embodied machines. We present four case studies of implementing a typical human-robot interaction scenario with different state-of-the-art dialog frameworks in order to identify challenges and pitfalls specific to HRI and potential solutions. The results are discussed with a special focus on the interplay between dialog and task modeling on robots.

We consider the computational complexity of probabilistic inference in Latent Dirichlet Allocation (LDA). First, we study the problem of finding the maximum a posteriori (MAP) assignment of topics to words, where the document's topic distribution is integrated out. We show that, when the effective number of topics per document is small, exact inference takes polynomial time. In contrast, we show that, when a document has a large number of topics, finding the MAP assignment of topics to words in LDA is NP-hard. Next, we consider the problem of finding the MAP topic distribution for a document, where the topic-word assignments are integrated out. We show that this problem is also NP-hard. Finally, we briefly discuss the problem of sampling from the posterior, showing that this is NP-hard in one restricted setting, but leaving open the general question.

Topic models have the potential to improve search and browsing by extracting useful semantic themes from web pages and other text documents. When learned topics are coherent and interpretable, they can be valuable for faceted browsing, results set diversity analysis, and document retrieval. However, when dealing with small collections or noisy text (e.g. web search result snippets or blog posts), learned topics can be less coherent, less interpretable, and less useful. To overcome this, we propose two methods to regularize the learning of topic models. Our regularizers work by creating a structured prior over words that reflect broad patterns in the external data. Using thirteen datasets we show that both regularizers improve topic coherence and interpretability while learning a faithful representation of the collection of interest. Overall, this work makes topic models more useful across a broader range of text data.

Topic models are learned via a statistical model of variation within document collections, but designed to extract meaningful semantic structure. Desirable traits include the ability to incorporate annotations or metadata associated with documents; the discovery of correlated patterns of topic usage; and the avoidance of parametric assumptions, such as manual specification of the number of topics. We propose a doubly correlated nonparametric topic (DCNT) model, the first model to simultaneously capture all three of these properties. The DCNT models metadata via a flexible, Gaussian regression on arbitrary input features; correlations via a scalable square-root covariance representation; and nonparametric selection from an unbounded series of potential topics via a stick-breaking construction. We validate the semantic structure and predictive performance of the DCNT using a corpus of NIPS documents annotated by various metadata.

Situated dialogue is prominent in the robot navigation task, where a human gives route instructions (i.e., a sequence of navigation commands) to an agent. We propose an approach for situated dialogue agents whereby they use strategies such as asking questions to repair or recover from unclear instructions, namely those that an agent misunderstands or considers ambiguous. Most immediately in this work we study examples from existing human-human dialogue corpora and relate them to our proposed approach.

Recent studies on question generation identify the need for automatic discourse analysers. We evaluated the feasibility of integrating an available discourse analyser called HILDA for a specific question generation system called CODA; introduce an approach by extracting a discourse corpus from the CODA parallel corpus; and identified future work towards automatic discourse analysis in the domain of question generation.

As people engage in increasingly complex conversations with computers, the need for generality and flexibility in spoken dialogue systems becomes more apparent. This pa­per describes how three different spoken dialogue systems for the same task reason with knowledge and certainty as they seek to understand what people want. It advocates sys­tems that exploit partial understanding, consider credibility, and are aware both of what they know and of their certainty that it matches their users’ intent.

Latent Dirichlet Allocation models discrete data as a mixture of discrete distributions, using Dirichlet beliefs over the mixture weights. We study a variation of this concept, in which the documents' mixture weight beliefs are replaced with squashed Gaussian distributions. This allows documents to be associated with elements of a Hilbert space, admitting kernel topic models (KTM), modelling temporal, spatial, hierarchical, social and other structure between documents. The main challenge is efficient approximate inference on the latent Gaussian. We present an approximate algorithm cast around a Laplace approximation in a transformed basis. The KTM can also be interpreted as a type of Gaussian process latent variable model, or as a topic model conditional on document features, uncovering links between earlier work in these areas.