Recently, there have been several high-profile achievements of agents learning to play games against humans and beat them. In this article, we study the problem of training intelligent agents in service of game development. Unlike the agents built to “beat the game,” our agents aim to produce human-like behavior to help with game evaluation and balancing. We discuss two fundamental metrics based on which we measure the human-likeness of agents, namely skill and style, which are multifaceted concepts with practical implications outlined in this article. We report four case studies in which the style and skill requirements inform the choice of algorithms and metrics used to train agents; ranging from A* search to state-of-the-art deep reinforcement learning (RL). Furthermore, we, show that the learning potential of state-of-the-art deep RL models does not seamlessly transfer from the benchmark environments to target ones without heavily tuning their hyperparameters, leading to linear scaling of the engineering efforts, and computational cost with the number of target domains.

Games have been a major focus of AI since the field formed seventy years ago. Recently, video games have replaced chess and go as the current "Mt. Everest Problem." This paper looks beyond the video games themselves to the application of AI techniques within the ecosystems that produce them. Electronic Arts (EA) must deal with AI at scale across many game studios as it develops many AAA games each year, and not a single, AI-based, flagship application. EA has adopted a horizontal scaling strategy in response to this challenge and built a platform for delivering AI artifacts anywhere within EA's software universe. By combining a data warehouse for player history, an Agent Store for capturing processes acquired through machine learning, and a recommendation engine as an action layer, EA has been delivering a wide range of AI solutions throughout the company during the last two years. These solutions, such as dynamic difficulty adjustment, in-game content and activity recommendations, matchmaking, and game balancing, have had major impact on engagement, revenue, and development resources within EA.

The digital game industry has recently adopted recommendation systems to provide suitable game and content choices to players. Recommendations in digital games have several unique applications and challenges compared to other well known recommendation system such as those for movies and books. Designers must adopt different architectures and algorithms to overcome these challenges. In this talk, we describe the game recommendation system at Electronic Arts. It leverages heterogeneous player data across many games to provide intelligent recommendations. We discuss three example applications: recommending games for purchase, suitable game map, and game difficulty.

Matchmaking connects multiple players to participate in online player-versus-player games. Current matchmaking systems depend on a single core strategy: create fair games at all times. These systems pair similarly skilled players on the assumption that a fair game is best player experience. We will demonstrate, however, that this intuitive assumption sometimes fails and that matchmaking based on fairness is not optimal for engagement.

In this paper, we present an efficient implementation of the fuzzy c-means clustering algorithm. The original algorithm alternates between estimating centers of the clusters and the fuzzy membership of the data points. The size of the membership matrix is on the order of the original data set, a prohibitive size if this technique is to be applied to very large data sets with many clusters. Our implementation eliminates the storage of this data structure by combining the two updates into a single update of the cluster centers. This change significantly affects the asymptotic runtime as the new algorithm is linear with respect to the number of clusters, while the original is quadratic. Elimination of the membership matrix also reduces the overhead associated with repeatedly accessing a large data structure. Empirical evidence is presented to quantify the savings achieved by this new method.

As the number of the processing units in a cluster computing system increases, what once were inconsequential details become major engineering concerns. Space, power distribution, and heat dissipation each demand attention for such clusters. We describe a simple rack mounting system that serves as a low-cost solution for clusters with fifty or more machines. Four motherboards of the same form factor are linked together with threaded rods and spacers and hang from inexpensive shelf brackets. This approach has enabled us to mount forty-eight motherboards and additional support hardware on a 6.5ft mobile rack with a 27in x 32in footprint for a fraction of the cost of traditional computer rack equipment or standard cases. In addition, the sharing of power supplies by eight motherboards has dramatically reduced complexity of our cluster. Efficient heat dissipation is demonstrated by thermal imaging.

This paper addresses the problem of integrating the effects of climate history and solar variability, to enhance regional hydrologic forecasting using neural networks. A previous attempt at modeling the inflow to Lake Okeechobee employed a multilayered perceptron (see Trimble et al, 1998). While the resulting model was able to capture some regularities of the measured inflow, it was far from being a useful predictive model. We continue the lake inflow modeling effort by examining data representation, quadratic input transformations, and time-delay neural networks.

Many researchers in AI and cognitive science believe that the complexity of a behavioural description reflects the underlying information processing complexity of the mechanism producing the behaviour. This paper explores the foundations of this complexity assumption. We first distinguish two types of complexity judgements that can be applied to these descriptions and then argue that neither type can be an intrinsic property of the underlying physical system. In short, we demonstrate how changes in the method of observation can radically alter both the number of apparent states and the apparent generative class of a system's behavioural description. From these examples we conclude that the act of observation can suggest frivolous computational explanations of physical phenomena, up to and including cognition.

Achilles cannot overtake a fleeing tortoise because in the interval of time that he takes to get to where the tortoise was, it can move away. But even if it should wait for him Achilles must first reach the half-way mark between them and he cannot do this unless he first reaches the half-way mark to that mark, and so on indefinitely. Against such an infinite conceptual regression he cannot even make a start, and so motion is impossible (Zeno, according to Zippin 1962, p. 10).

Many connectionist approaches to musical expectancy and music composition let the question of ‘What next?’ overshadow the equally important question of ‘When next?’. One cannot escape the latter question, one of temporal structure, when considering the perception of musical meter. We view the perception of metrical structure as a dynamic process where the temporal organization of external musical events synchronizes, or entrains, a listener's internal processing mechanisms. This article introduces a novel connectionist unit, based upon a mathematical model of entrainment, capable of phase- and frequency-locking to periodic components of incoming rhythmic patterns. Networks of these units can self-organize temporally structured responses to rhythmic patterns. The resulting network behavior embodies the perception of metrical structure. The article concludes with a discussion of the implications of our approach for theories of metrical structure and musical expectancy.

A set of experiments that precisely identify the power and limitations of the method of back-propagation is reported. The experiment on learning to compute the exclusive-OR function suggests that the computational efficiency of learning by the method of back-propagation depends on the initial weights in the network. The experiment on learning to play tic-tac-toe suggests that the information content of what is learned by the back-propagation method is dependent on the initial abstractions in the network. It also suggests that these abstractions are a major source of power for learning in parallel distributed processing networks. In addition, it is shown that the learning task addressed by connectionist methods, including the back-propagation method, is computationally intractable. These experimental and theoretical results strongly indicate that current connectionist methods may be too limited for the complex task of learning they seek to solve. It is proposed that the power of neural networks may be enhanced by developing task-specific connectionist methods.>