We present and study a partial-information model of online learning, where a decision maker repeatedly chooses from a finite set of actions, and observes some subset of the associated losses. This naturally models several situations where the losses of different actions are related, and knowing the loss of one action provides information on the loss of other actions. Moreover, it generalizes and interpolates between the well studied full-information setting (where all losses are revealed) and the bandit setting (where only the loss of the action chosen by the player is revealed). We provide several algorithms addressing different variants of our setting, and provide tight regret bounds depending on combinatorial properties of the information feedback structure.

Recent advances in the estimation of deep directed graphical models and recurrent networks let us contribute to the removal of a blind spot in the area of probabilistc modelling of time series. The proposed methods i) can infer distributed latent state-space trajectories with nonlinear transitions, ii) scale to large data sets thanks to the use of a stochastic objective and fast, approximate inference, iii) enable the design of rich emission models which iv) will naturally lead to structured outputs. Two different paths of introducing latent state sequences are pursued, leading to the variational recurrent auto encoder (VRAE) and the variational one step predictor (VOSP). The use of independent Wiener processes as priors on the latent state sequence is a viable compromise between efficient computation of the Kullback-Leibler divergence from the variational approximation of the posterior and maintaining a reasonable belief in the dynamics. We verify our methods empirically, obtaining results close or superior to the state of the art. We also show qualitative results for denoising and missing value imputation.

A Survey of Artificial Intelligence Research at the IIIA

The algorithm selection problem is concerned with selecting the best algorithm to solve a given problem instance on a case-by-case basis. It has become especially relevant in the last decade, with researchers increasingly investigating how to identify the most suitable existing algorithm for solving a problem instance instead of developing new algorithms. This survey presents an overview of this work focusing on the contributions made in the area of combinatorial search problems, where algorithm selection techniques have achieved significant performance improvements. We unify and organise the vast literature according to criteria that determine algorithm selection systems in practice.

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.

Computational sustainability problems, which exist in dynamic environments with high amounts of uncertainty, provide a variety of unique challenges to artificial intelligence research and the opportunity for significant impact upon our collective future. This editorial introduction provides an overview of artificial intelligence for computational sustainability, and introduces the special issue articles that appear in this issue and the previous issue of AI Magazine.

Reinforcement learning is one of the most general problems in artificial intelligence. It has been used to model problems in automated experiment design, control, economics, game playing, scheduling and telecommunications. The aim of the reinforcement learning competition is to encourage the development of very general learning agents for arbitrary reinforcement learning problems and to provide a test-bed for the unbiased evaluation of algorithms.

While there have been many papers published on path planning in grids, there has not been significant work on comparing existing approaches, and it is difficult to evaluate new work in comparison to existing work. After creating a public repository of grid-based path planning problems we created the grid-based planning competition (GPPC) to facilitate these comparisons. This article describes the motivation and design of the competition, as well as plans for the future of the competition.

Recent developments in linear system identification have proposed the use of non-parameteric methods, relying on regularization strategies, to handle the so-called bias/variance trade-off. This paper introduces an impulse response estimator which relies on an $\ell_2$-type regularization including a rank-penalty derived using the log-det heuristic as a smooth approximation to the rank function. This allows to account for different properties of the estimated impulse response (e.g. smoothness and stability) while also penalizing high-complexity models. This also allows to account and enforce coupling between different input-output channels in MIMO systems. According to the Bayesian paradigm, the parameters defining the relative weight of the two regularization terms as well as the structure of the rank penalty are estimated optimizing the marginal likelihood. Once these hyperameters have been estimated, the impulse response estimate is available in closed form. Experiments show that the proposed method is superior to the estimator relying on the "classic" $\ell_2$-regularization alone as well as those based in atomic and nuclear norm.

Gaussian processes (GPs) are a powerful tool for probabilistic inference over functions. They have been applied to both regression and non-linear dimensionality reduction, and offer desirable properties such as uncertainty estimates, robustness to over-fitting, and principled ways for tuning hyper-parameters. However the scalability of these models to big datasets remains an active topic of research. We introduce a novel re-parametrisation of variational inference for sparse GP regression and latent variable models that allows for an efficient distributed algorithm. This is done by exploiting the decoupling of the data given the inducing points to re-formulate the evidence lower bound in a Map-Reduce setting. We show that the inference scales well with data and computational resources, while preserving a balanced distribution of the load among the nodes. We further demonstrate the utility in scaling Gaussian processes to big data. We show that GP performance improves with increasing amounts of data in regression (on flight data with 2 million records) and latent variable modelling (on MNIST). The results show that GPs perform better than many common models often used for big data.