Recent years have witnessed growing interest in data-driven approaches to interactive narrative planning and drama management. Reinforcement learning techniques show particular promise because they can automatically induce and refine models for tailoring game events by optimizing reward functions that explicitly encode interactive narrative experiences' quality. Due to the inherently subjective nature of interactive narrative experience, designing effective reward functions is challenging. In this paper, we investigate the impacts of alternate formulations of reward in a reinforcement learning-based interactive narrative planner for the Crystal Island game environment.

Automatic analysis of game levels can provide as- sistance to game designers and procedural content generation. We introduce a static-dynamic scale to categorize level analysis strategies, which captures the extent that the analysis depends on player simulation. Due to its ability to automatically learn intermediate representations for the task, a convolutional neural network (CNN) provides a general tool for both types of analysis. In this paper, we explore the use of CNN to analyze 1,437 Infinite Mario levels. We further propose a deep reinforcement learning technique for dynamic analysis, which allows the simulated player to pay a penalty to reduce error in its control. We empirically demonstrate the effectiveness of our techniques and complementarity of dynamic and static analysis.

Real-Time Strategy games have become a popular test-bed for modern AI system due to their real-time computational constraints, complex multi-unit control problems, and imperfect information. One of the most important aspects of any RTS AI system is the efficient control of units in complex combat scenarios, also known as micromanagement. Recently, a model-based heuristic search technique called Portfolio Greedy Search (PGS) has shown promisingpaper we present the first integration of PGS into the StarCraft game engine, and compare its performance to the current state-of-the-art deep reinforcement learning method in several benchmark combat scenarios. We then perform theperformance for providing real-time decision making in RTS combat scenarios, but has so far only been tested in SparCraft: an RTS combat simulator. In this same experiments within the SparCraft simulator in order to investigate any differences between PGS performance in the simulator and in the actual game. Lastly, we investigate how varying parameters of the SparCraft simulator affect the performance of PGS in the StarCraft game engine. We demonstrate that the performance of PGS relies heavily on the accuracy of the underlying model, outperforming other techniques only for scenarios where the SparCraft simulation model more accurately matches the StarCraft game engine.

We release a dataset of 65646 StarCraft replays that contains 1535 million frames and 496 million player actions. We provide full game state data along with the original replays that can be viewed in StarCraft. The game state data was recorded every 3 frames which ensures suitability for a wide variety of machine learning tasks such as strategy classification, inverse reinforcement learning, imitation learning, forward modeling, partial information extraction, and others. We use TorchCraft to extract and store the data, which standardizes the data format for both reading from replays and reading directly from the game. Furthermore, the data can be used on different operating systems and platforms. The dataset contains valid, non-corrupted replays only and its quality and diversity was ensured by a number of heuristics. We illustrate the diversity of the data with various statistics and provide examples of tasks that benefit from the dataset.

We present a novel modular architecture for StarCraft II AI. The architecture splits responsibilities between multiple modules that each control one aspect of the game, such as buildorder selection or tactics. A centralized scheduler reviews macros suggested by all modules and decides their order of execution. An updater keeps track of environment changes and instantiates macros into series of executable actions. Modules in this framework can be optimized independently or jointly via human design, planning, or reinforcement learning. We present the first result of applying deep reinforcement learning techniques to training a modular agent with selfplay, achieving 92% or 86% win rates against the "Harder" (level 5) built-in Blizzard bot in Zerg vs. Zerg matches, with or without fog-of-war.

Demand response (DR), as one of the important energy resources in the future's grid, provides the services of peak shaving, enhancing the efficiency of renewable energy utilization with a short response period, and low cost. Various categories of DR are established, e.g. automated DR, incentive DR, emergency DR, and demand bidding. However, with the practical issue of the unawareness of residential and commercial consumers' utility models, the researches about demand bidding aggregator involved in the electricity market are just at the beginning stage. For this issue, the bidding price and bidding quantity are two required decision variables while considering the uncertainties due to the market and participants. In this paper, we determine the bidding and purchasing strategy simultaneously employing the smart meter data and functions. A two-agent deep deterministic policy gradient method is developed to optimize the decisions through learning historical bidding experiences. The online learning further utilizes the daily newest bidding experience attained to ensure trend tracing and self-adaptation. Two environment simulators are adopted for testifying the robustness of the model. The results prove that when facing diverse situations the proposed model can earn the optimal profit via off/online learning the bidding rules and robustly making the proper bid.

Real-world sequential decision making problems commonly involve partial observability, which requires the agent to maintain a memory of history in order to infer the latent states, plan and make good decisions. Coping with partial observability in general is extremely challenging, as a number of worst-case statistical and computational barriers are known in learning Partially Observable Markov Decision Processes (POMDPs). Motivated by the problem structure in several physical applications, as well as a commonly used technique known as "frame stacking", this paper proposes to study a new subclass of POMDPs, whose latent states can be decoded by the most recent history of a short length $m$. We establish a set of upper and lower bounds on the sample complexity for learning near-optimal policies for this class of problems in both tabular and rich-observation settings (where the number of observations is enormous). In particular, in the rich-observation setting, we develop new algorithms using a novel "moment matching" approach with a sample complexity that scales exponentially with the short length $m$ rather than the problem horizon, and is independent of the number of observations. Our results show that a short-term memory suffices for reinforcement learning in these environments.

We propose a new policy parameterization for representing 3D rotations during reinforcement learning. Today in the continuous control reinforcement learning literature, many stochastic policy parameterizations are Gaussian. We argue that universally applying a Gaussian policy parameterization is not always desirable for all environments. One such case in particular where this is true are tasks that involve predicting a 3D rotation output, either in isolation, or coupled with translation as part of a full 6D pose output. Our proposed Bingham Policy Parameterization (BPP) models the Bingham distribution and allows for better rotation (quaternion) prediction over a Gaussian policy parameterization in a range of reinforcement learning tasks. We evaluate BPP on the rotation Wahba problem task, as well as a set of vision-based next-best pose robot manipulation tasks from RLBench. We hope that this paper encourages more research into developing other policy parameterization that are more suited for particular environments, rather than always assuming Gaussian.

skrl is an open-source modular library for reinforcement learning written in Python and designed with a focus on readability, simplicity, and transparency of algorithm implementations. Apart from supporting environments that use the traditional OpenAI Gym interface, it allows loading, configuring, and operating NVIDIA Isaac Gym environments, enabling the parallel training of several agents with adjustable scopes, which may or may not share resources, in the same execution. The library's documentation can be found at https://skrl.readthedocs.io and its source code is available on GitHub at url{https://github.com/Toni-SM/skrl.

Consider a walking agent that must adapt to damage. To approach this task, we can train a collection of policies and have the agent select a suitable policy when damaged. Training this collection may be viewed as a quality diversity (QD) optimization problem, where we search for solutions (policies) which maximize an objective (walking forward) while spanning a set of measures (measurable characteristics). Recent work shows that differentiable quality diversity (DQD) algorithms greatly accelerate QD optimization when exact gradients are available for the objective and measures. However, such gradients are typically unavailable in RL settings due to non-differentiable environments. To apply DQD in RL settings, we propose to approximate objective and measure gradients with evolution strategies and actor-critic methods. We develop two variants of the DQD algorithm CMA-MEGA, each with different gradient approximations, and evaluate them on four simulated walking tasks. One variant achieves comparable performance (QD score) with the state-of-the-art PGA-MAP-Elites in two tasks. The other variant performs comparably in all tasks but is less efficient than PGA-MAP-Elites in two tasks. These results provide insight into the limitations of CMA-MEGA in domains that require rigorous optimization of the objective and where exact gradients are unavailable.