The aim of the Disco: Human and Machine Learning in Games workshop was to extend upon the focus of two past workshops and explore the intersection of entertainment, learning and human computation. The goal of the workshop was to examine both human learning and machine learning in games and human computation. Human computation methods let machines learn from humans where games can provide humans the opportunity to learn. The workshop was thus devoted to I learn, in Latin disco, for machines and humans alike. The First AAAI Conference on Human Computation and Crowdsourcing Was Held in the Southern California Desert Community of Palm Springs.

The workshop was accompanied by an evening Games are unique in that their components event, DAGGER, which drew together local game developers (from the rules and goals of the game to the appearance and academic research projects. Acting both of avatars and their dialogue) must encompass as an exhibition and as an informal gathering, the both functional and aesthetic prerequisites. Artificial DAGGER event allowed attendees to interact directly intelligence usually focuses on the functional quality with a wide variety of game types and technologies, of such game components, for example, ensuring as well as with their developers. As events such that an avatar can traverse a level in minimal time or as DAGGER help bridge the gap between theoretical that AI can win over any human in a strategy game. The papers avatar, or level would appeal to a particular player. of the workshop were published as AAAI Technical The Workshop on AI and Game Aesthetics provided Report WS-13-19.

MiniZinc is a solver agnostic modeling language for defining and solver combinatorial satisfaction and optimization problems. MiniZinc provides a solver independent modeling language which is now supported by constraint programming solvers, mixed integer programming solvers, SAT and SAT modulo theory solvers, and hybrid solvers. Since 2008 we have run the MiniZinc challenge every year, which compares and contrasts the different strengths of different solvers and solving technologies on a set of MiniZinc models. Here we report on what we have learnt from running the competition for 6 years.

Many problems in computational sustainability require making a sequence of decisions in complex, uncertain environments. Such problems are generally notoriously difficult. In this article, we review the recently discovered notion of adaptive submodularity, an intuitive diminishing returns condition that generalizes the classical notion of submodular set functions to sequential decision problems. Problems exhibiting the adaptive submodularity property can be efficiently and provably near-optimally solved using simple myopic policies. We illustrate this concept in several case studies of interest in computational sustainability: First, we demonstrate how it can be used to efficiently plan for resolving uncertainty in adaptive management scenarios. Secondly, we show how it applies to dynamic conservation planning for protecting endangered species, a case study carried out in collaboration with the US Geological Survey and the US Fish and Wildlife Service.

Ecological systems are inherently complex. The processes that affect the distributions of animals and plants operate at multiple spatial and temporal scales, presenting a unique challenge for the development and coordination of effective conservation strategies, particularly for wide-ranging species. In order to study ecological systems across scales, data must be collected at fine resolutions across broad spatial and temporal extents. Crowdsourcing has emerged as an efficient way to gather these data by engaging large numbers of people to record observations. However, data gathered by crowdsourced projects are often biased due to the opportunistic approach of data collection. In this article, we propose a general class of models called AdaSTEM, (for adaptive spatio-temporal exploratory models), that are designed to meet these challenges by adapting to multiple scales while exploiting variation in data density common with crowdsourced data. To illustrate the use of AdaSTEM, we produce intra-seasonal distribution estimates of long-distance migrations across the Western Hemisphere using data from eBird, a citizen science project that utilizes volunteers to collect observations of birds. Subsequently, model diagnostics are used to quantify and visualize the scale and quality of distribution estimates. This analysis shows how AdaSTEM can automatically adapt to complex spatiotemporal processes across a range of scales, thus providing essential information for full-life cycle conservation planning of broadly distributed species, communities, and ecosystems.

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 provides an overview of artificial intelligence for computational sustainability, and introduces this special issue of AI Magazine.

Our goal is to build robust optimization problems for making decisions based on complex data from the past. In robust optimization (RO) generally, the goal is to create a policy for decision-making that is robust to our uncertainty about the future. In particular, we want our policy to best handle the the worst possible situation that could arise, out of an uncertainty set of possible situations. Classically, the uncertainty set is simply chosen by the user, or it might be estimated in overly simplistic ways with strong assumptions; whereas in this work, we learn the uncertainty set from data collected in the past. The past data are drawn randomly from an (unknown) possibly complicated high-dimensional distribution. We propose a new uncertainty set design and show how tools from statistical learning theory can be employed to provide probabilistic guarantees on the robustness of the policy.

Recommendation systems have been integrated into the majority of large online systems. They tailor those systems to individual users by filtering and ranking information according to user profiles. This adaptation process influences the way users interact with the system and, as a consequence, increases the difficulty of evaluating a recommendation algorithm with historical data (via offline evaluation). This paper analyses this evaluation bias and proposes a simple item weighting solution that reduces its impact. The efficiency of the proposed solution is evaluated on real world data extracted from Viadeo professional social network.

AUC (area under ROC curve) is an important evaluation criterion, which has been popularly used in many learning tasks such as class-imbalance learning, cost-sensitive learning, learning to rank, etc. Many learning approaches try to optimize AUC, while owing to the non-convexity and discontinuousness of AUC, almost all approaches work with surrogate loss functions. Thus, the consistency of AUC is crucial; however, it has been almost untouched before. In this paper, we provide a sufficient condition for the asymptotic consistency of learning approaches based on surrogate loss functions. Based on this result, we prove that exponential loss and logistic loss are consistent with AUC, but hinge loss is inconsistent. Then, we derive the $q$-norm hinge loss and general hinge loss that are consistent with AUC. We also derive the consistent bounds for exponential loss and logistic loss, and obtain the consistent bounds for many surrogate loss functions under the non-noise setting. Further, we disclose an equivalence between the exponential surrogate loss of AUC and exponential surrogate loss of accuracy, and one straightforward consequence of such finding is that AdaBoost and RankBoost are equivalent.

A successful approach to structured learning is to write the learning objective as a joint function of linear parameters and inference messages, and iterate between updates to each. This paper observes that if the inference problem is "smoothed" through the addition of entropy terms, for fixed messages, the learning objective reduces to a traditional (non-structured) logistic regression problem with respect to parameters. In these logistic regression problems, each training example has a bias term determined by the current set of messages. Based on this insight, the structured energy function can be extended from linear factors to any function class where an "oracle" exists to minimize a logistic loss.