Collaborative filtering (CF) aims to predict users' ratings on items according to historical user-item preference data. In many real-world applications, preference data are usually sparse, which would make models overfit and fail to give accurate predictions. Recently, several research works show that by transferring knowledge from some manually selected source domains, the data sparseness problem could be mitigated. However for most cases, parts of source domain data are not consistent with the observations in the target domain, which may misguide the target domain model building. In this paper, we propose a novel criterion based on empirical prediction error and its variance to better capture the consistency across domains in CF settings. Consequently, we embed this criterion into a boosting framework to perform selective knowledge transfer. Comparing to several state-of-the-art methods, we show that our proposed selective transfer learning framework can significantly improve the accuracy of rating prediction tasks on several real-world recommendation tasks.

The hidden Markov model (HMM) is a widely-used generative model that copes with sequential data, assuming that each observation is conditioned on the state of a hidden Markov chain. In this paper, we derive a novel algorithm to cluster HMMs based on the hierarchical EM (HEM) algorithm. The proposed algorithm i) clusters a given collection of HMMs into groups of HMMs that are similar, in terms of the distributions they represent, and ii) characterizes each group by a "cluster center", i.e., a novel HMM that is representative for the group, in a manner that is consistent with the underlying generative model of the HMM. To cope with intractable inference in the E-step, the HEM algorithm is formulated as a variational optimization problem, and efficiently solved for the HMM case by leveraging an appropriate variational approximation. The benefits of the proposed algorithm, which we call variational HEM (VHEM), are demonstrated on several tasks involving time-series data, such as hierarchical clustering of motion capture sequences, and automatic annotation and retrieval of music and of online hand-writing data, showing improvements over current methods. In particular, our variational HEM algorithm effectively leverages large amounts of data when learning annotation models by using an efficient hierarchical estimation procedure, which reduces learning times and memory requirements, while improving model robustness through better regularization.

Many latent (factorized) models have been proposed for recommendation tasks like collaborative filtering and for ranking tasks like document or image retrieval and annotation. Common to all those methods is that during inference the items are scored independently by their similarity to the query in the latent embedding space. The structure of the ranked list (i.e. considering the set of items returned as a whole) is not taken into account. This can be a problem because the set of top predictions can be either too diverse (contain results that contradict each other) or are not diverse enough. In this paper we introduce a method for learning latent structured rankings that improves over existing methods by providing the right blend of predictions at the top of the ranked list. Particular emphasis is put on making this method scalable. Empirical results on large scale image annotation and music recommendation tasks show improvements over existing approaches.

Previous work on recommender systems mainly focus on fitting the ratings provided by users. However, the response patterns, i.e., some items are rated while others not, are generally ignored. We argue that failing to observe such response patterns can lead to biased parameter estimation and sub-optimal model performance. Although several pieces of work have tried to model users' response patterns, they miss the effectiveness and interpretability of the successful matrix factorization collaborative filtering approaches. To bridge the gap, in this paper, we unify explicit response models and PMF to establish the Response Aware Probabilistic Matrix Factorization (RAPMF) framework. We show that RAPMF subsumes PMF as a special case. Empirically we demonstrate the merits of RAPMF from various aspects.

This paper considers stochastic bandits with side observations, a model that accounts for both the exploration/exploitation dilemma and relationships between arms. In this setting, after pulling an arm i, the decision maker also observes the rewards for some other actions related to i. We will see that this model is suited to content recommendation in social networks, where users' reactions may be endorsed or not by their friends. We provide efficient algorithms based on upper confidence bounds (UCBs) to leverage this additional information and derive new bounds improving on standard regret guarantees. We also evaluate these policies in the context of movie recommendation in social networks: experiments on real datasets show substantial learning rate speedups ranging from 2.2x to 14x on dense networks.

Automated sports commentary is a form of automated narrative. Sports commentary exists to keep the viewer informed and entertained. One way to entertain the viewer is by telling brief stories relevant to the game in progress. We introduce a system called the Sports Commentary Recommendation System (SCoReS) that can automatically suggest stories for commentators to tell during games. Through several user studies, we compared commentary using SCoReS to three other types of commentary and show that SCoReS adds significantly to the broadcast across several enjoyment metrics. We also collected interview data from professional sports commentators who positively evaluated a demonstration of the system. We conclude that SCoReS can be a useful broadcast tool, effective at selecting stories that add to the enjoyment and watchability of sports. SCoReS is a step toward automating sports commentary and, thus, automating narrative.

Autonomous aesthetic evaluation is the Holy Grail of generative music, and one of the great challenges of computational creativity. Unlike most other computational activities, there is no notion of optimality in evaluating creative output: there are subjective impressions involved, and framing obviously plays a big role. When developing metacreative systems, a purely objective fitness function is not available: the designer is thus faced with how much of their own aesthetic to include. Can a generative system be free of the designer’s bias? This paper presents a system that incorporates an aesthetic selection process that allows for both human-designed and non-human fitness functions.

We describe ANGELINA3, a system that can automatically develop games along a defined theme, by selecting appropriate multimedia content from a variety of sources and incorporating it into a game's design. We discuss these capabilities in the context of the FACE model for assessing progress in the building of creative systems, and discuss how ANGELINA3 can be improved through further work.

This demonstration presents MaestroGenesis, a program that helps users create complete polyphonic musical pieces from as little as a simple, human composed monophonic melody. MaestroGenesis creates music by exploiting two key ideas behind the functional scaffolding for musical composition (FSMC) approach: (1) that music a function of time and (2) that functional transformations of initial human starting melodies, or scaffolds, inherit some of the essential human qualities contained in the scaffold. Music in FSMC is represented as a functional relationship between the scaffold and a generated accompaniment. The GUI helps users evolve these functions by importing and developing their music through a breeding process akin to animal breeding, called interactive evolutionary computation. Some resulting pieces are indistinguishable from completely human-composed pieces.

Most systems for generating cinematic shot sequences for virtual environments focus on the low-level problems of camera placement. While this approach will create a sequence of camera shots which film individual events in a virtual environment, it does not account for the high-level effects shot sequences have on viewer inferences. There are systems which are based on well known cinematography principles such as the rule of thirds and other framing principals, however these usually utilize schemas or predefined shots and do not reason about the high level cognitive effects on the viewer. In this paper a system is proposed which can reason directly about these high-level cognitive and narrative effects of a shot sequence on the viewer’s mental state.