Collaborating Authors


School of hard knocks: Curriculum analysis for Pommerman with a fixed computational budget Artificial Intelligence

Pommerman is a hybrid cooperative/adversarial multi-agent environment, with challenging characteristics in terms of partial observability, limited or no communication, sparse and delayed rewards, and restrictive computational time limits. This makes it a challenging environment for reinforcement learning (RL) approaches. In this paper, we focus on developing a curriculum for learning a robust and promising policy in a constrained computational budget of 100,000 games, starting from a fixed base policy (which is itself trained to imitate a noisy expert policy). All RL algorithms starting from the base policy use vanilla proximal-policy optimization (PPO) with the same reward function, and the only difference between their training is the mix and sequence of opponent policies. One expects that beginning training with simpler opponents and then gradually increasing the opponent difficulty will facilitate faster learning, leading to more robust policies compared against a baseline where all available opponent policies are introduced from the start. We test this hypothesis and show that within constrained computational budgets, it is in fact better to "learn in the school of hard knocks", i.e., against all available opponent policies nearly from the start. We also include ablation studies where we study the effect of modifying the base environment properties of ammo and bomb blast strength on the agent performance.

Applied Machine Learning for Games: A Graduate School Course Artificial Intelligence

The game industry is moving into an era where old-style game engines are being replaced by re-engineered systems with embedded machine learning technologies for the operation, analysis and understanding of game play. In this paper, we describe our machine learning course designed for graduate students interested in applying recent advances of deep learning and reinforcement learning towards gaming. This course serves as a bridge to foster interdisciplinary collaboration among graduate schools and does not require prior experience designing or building games. Graduate students enrolled in this course apply different fields of machine learning techniques such as computer vision, natural language processing, computer graphics, human computer interaction, robotics and data analysis to solve open challenges in gaming. Student projects cover use-cases such as training AI-bots in gaming benchmark environments and competitions, understanding human decision patterns in gaming, and creating intelligent non-playable characters or environments to foster engaging gameplay. Projects demos can help students open doors for an industry career, aim for publications, or lay the foundations of a future product. Our students gained hands-on experience in applying state of the art machine learning techniques to solve real-life problems in gaming.

Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge? Artificial Intelligence

Most recently developed approaches to cooperative multi-agent reinforcement learning in the \emph{centralized training with decentralized execution} setting involve estimating a centralized, joint value function. In this paper, we demonstrate that, despite its various theoretical shortcomings, Independent PPO (IPPO), a form of independent learning in which each agent simply estimates its local value function, can perform just as well as or better than state-of-the-art joint learning approaches on popular multi-agent benchmark suite SMAC with little hyperparameter tuning. We also compare IPPO to several variants; the results suggest that IPPO's strong performance may be due to its robustness to some forms of environment non-stationarity.

Instance based Generalization in Reinforcement Learning Machine Learning

Agents trained via deep reinforcement learning (RL) routinely fail to generalize to unseen environments, even when these share the same underlying dynamics as the training levels. Understanding the generalization properties of RL is one of the challenges of modern machine learning. Towards this goal, we analyze policy learning in the context of Partially Observable Markov Decision Processes (POMDPs) and formalize the dynamics of training levels as instances. We prove that, independently of the exploration strategy, reusing instances introduces significant changes on the effective Markov dynamics the agent observes during training. Maximizing expected rewards impacts the learned belief state of the agent by inducing undesired instance-specific speed-running policies instead of generalizable ones, which are sub-optimal on the training set. We provide generalization bounds to the value gap in train and test environments based on the number of training instances, and use insights based on these to improve performance on unseen levels. We propose training a shared belief representation over an ensemble of specialized policies, from which we compute a consensus policy that is used for data collection, disallowing instance-specific exploitation. We experimentally validate our theory, observations, and the proposed computational solution over the CoinRun benchmark.

An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective Artificial Intelligence

Following the remarkable success of the AlphaGO series, 2019 was a booming year that witnessed significant advances in multi-agent reinforcement learning (MARL) techniques. MARL corresponds to the learning problem in a multi-agent system in which multiple agents learn simultaneously. MARL is an interdisciplinary domain with a long history that includes game theory, machine learning, stochastic control, psychology, and optimisation. Although MARL has achieved considerable empirical success in solving real-world games, there is a lack of a self-contained overview in the literature that elaborates the game theoretical foundations of modern MARL methods and summarises the recent advances. In fact, the majority of existing surveys are outdated and do not fully cover the recent developments since 2010. In this work, we provide a monograph on MARL that covers both the fundamentals and the latest developments in the research frontier. The goal of our monograph is to provide a self-contained assessment of the current state-of-the-art MARL techniques from a game theoretical perspective. We expect this work to serve as a stepping stone for both new researchers who are about to enter this fast-growing domain and existing domain experts who want to obtain a panoramic view and identify new directions based on recent advances.

Align-RUDDER: Learning From Few Demonstrations by Reward Redistribution Artificial Intelligence

Reinforcement Learning algorithms require a large number of samples to solve complex tasks with sparse and delayed rewards. Complex tasks can often be hierarchically decomposed into sub-tasks. A step in the Q-function can be associated with solving a sub-task, where the expectation of the return increases. RUDDER has been introduced to identify these steps and then redistribute reward to them, thus immediately giving reward if sub-tasks are solved. Since the problem of delayed rewards is mitigated, learning is considerably sped up. However, for complex tasks, current exploration strategies as deployed in RUDDER struggle with discovering episodes with high rewards. Therefore, we assume that episodes with high rewards are given as demonstrations and do not have to be discovered by exploration. Typically the number of demonstrations is small and RUDDER's LSTM model as a deep learning method does not learn well. Hence, we introduce Align-RUDDER, which is RUDDER with two major modifications. First, Align-RUDDER assumes that episodes with high rewards are given as demonstrations, replacing RUDDER's safe exploration and lessons replay buffer. Second, we replace RUDDER's LSTM model by a profile model that is obtained from multiple sequence alignment of demonstrations. Profile models can be constructed from as few as two demonstrations as known from bioinformatics. Align-RUDDER inherits the concept of reward redistribution, which considerably reduces the delay of rewards, thus speeding up learning. Align-RUDDER outperforms competitors on complex artificial tasks with delayed reward and few demonstrations. On the MineCraft ObtainDiamond task, Align-RUDDER is able to mine a diamond, though not frequently. Github:, YouTube:

The Bottleneck Simulator: A Model-Based Deep Reinforcement Learning Approach

Journal of Artificial Intelligence Research

Deep reinforcement learning has recently shown many impressive successes. However, one major obstacle towards applying such methods to real-world problems is their lack of data-efficiency. To this end, we propose the Bottleneck Simulator: a model-based reinforcement learning method which combines a learned, factorized transition model of the environment with rollout simulations to learn an effective policy from few examples. The learned transition model employs an abstract, discrete (bottleneck) state, which increases sample efficiency by reducing the number of model parameters and by exploiting structural properties of the environment. We provide a mathematical analysis of the Bottleneck Simulator in terms of fixed points of the learned policy, which reveals how performance is affected by four distinct sources of error: an error related to the abstract space structure, an error related to the transition model estimation variance, an error related to the transition model estimation bias, and an error related to the transition model class bias. Finally, we evaluate the Bottleneck Simulator on two natural language processing tasks: a text adventure game and a real-world, complex dialogue response selection task.

Human-in-the-Loop Methods for Data-Driven and Reinforcement Learning Systems Artificial Intelligence

Recent successes combine reinforcement learning algorithms and deep neural networks, despite reinforcement learning not being widely applied to robotics and real world scenarios. This can be attributed to the fact that current state-of-the-art, end-to-end reinforcement learning approaches still require thousands or millions of data samples to converge to a satisfactory policy and are subject to catastrophic failures during training. Conversely, in real world scenarios and after just a few data samples, humans are able to either provide demonstrations of the task, intervene to prevent catastrophic actions, or simply evaluate if the policy is performing correctly. This research investigates how to integrate these human interaction modalities to the reinforcement learning loop, increasing sample efficiency and enabling real-time reinforcement learning in robotics and real world scenarios. This novel theoretical foundation is called Cycle-of-Learning, a reference to how different human interaction modalities, namely, task demonstration, intervention, and evaluation, are cycled and combined to reinforcement learning algorithms. Results presented in this work show that the reward signal that is learned based upon human interaction accelerates the rate of learning of reinforcement learning algorithms and that learning from a combination of human demonstrations and interventions is faster and more sample efficient when compared to traditional supervised learning algorithms. Finally, Cycle-of-Learning develops an effective transition between policies learned using human demonstrations and interventions to reinforcement learning. The theoretical foundation developed by this research opens new research paths to human-agent teaming scenarios where autonomous agents are able to learn from human teammates and adapt to mission performance metrics in real-time and in real world scenarios.

Model-based Reinforcement Learning: A Survey Artificial Intelligence

Sequential decision making, commonly formalized as Markov Decision Process (MDP) optimization, is a key challenge in artificial intelligence. Two key approaches to this problem are reinforcement learning (RL) and planning. This paper presents a survey of the integration of both fields, better known as model-based reinforcement learning. Model-based RL has two main steps. First, we systematically cover approaches to dynamics model learning, including challenges like dealing with stochasticity, uncertainty, partial observability, and temporal abstraction. Second, we present a systematic categorization of planning-learning integration, including aspects like: where to start planning, what budgets to allocate to planning and real data collection, how to plan, and how to integrate planning in the learning and acting loop. After these two key sections, we also discuss the potential benefits of model-based RL, like enhanced data efficiency, targeted exploration, and improved stability. Along the survey, we also draw connections to several related RL fields, like hierarchical RL and transfer, and other research disciplines, like behavioural psychology. Altogether, the survey presents a broad conceptual overview of planning-learning combinations for MDP optimization.

Systematic Generalisation through Task Temporal Logic and Deep Reinforcement Learning Artificial Intelligence

This paper presents a neuro-symbolic agent that combines deep reinforcement learning (DRL) with temporal logic (TL), and achieves systematic out-of-distribution generalisation in tasks that involve following a formally specified instruction. Specifically, the agent learns general notions of negation and disjunction, and successfully applies them to previously unseen objects without further training. To this end, we also introduce Task Temporal Logic (TTL), a learning-oriented formal language, whose atoms are designed to help the training of a DRL agent targeting systematic generalisation. To validate this combination of logic-based and neural-network techniques, we provide experimental evidence for the kind of neural-network architecture that most enhances the generalisation performance of the agent. Our findings suggest that the right architecture can significatively improve the ability of the agent to generalise in systematic ways, even with abstract operators, such as negation, which previous research have struggled with.