The last decade has seen a significant increase of interest in deep learning research, with many public successes that have demonstrated its potential. As such, these systems are now being incorporated into commercial products. With this comes an additional challenge: how can we build AI systems that solve tasks where there is not a crisp, well-defined specification? While multiple solutions have been proposed, in this competition we focus on one in particular: learning from human feedback. Rather than training AI systems using a predefined reward function or using a labeled dataset with a predefined set of categories, we instead train the AI system using a learning signal derived from some form of human feedback, which can evolve over time as the understanding of the task changes, or as the capabilities of the AI system improve. The MineRL BASALT competition aims to spur forward research on this important class of techniques. We design a suite of four tasks in Minecraft for which we expect it will be hard to write down hardcoded reward functions. These tasks are defined by a paragraph of natural language: for example, "create a waterfall and take a scenic picture of it", with additional clarifying details. Participants must train a separate agent for each task, using any method they want. Agents are then evaluated by humans who have read the task description. To help participants get started, we provide a dataset of human demonstrations on each of the four tasks, as well as an imitation learning baseline that leverages these demonstrations. Our hope is that this competition will improve our ability to build AI systems that do what their designers intend them to do, even when the intent cannot be easily formalized. Besides allowing AI to solve more tasks, this can also enable more effective regulation of AI systems, as well as making progress on the value alignment problem.

Reinforcement learning competitions have formed the basis for standard research benchmarks, galvanized advances in the state-of-the-art, and shaped the direction of the field. Despite this, a majority of challenges suffer from the same fundamental problems: participant solutions to the posed challenge are usually domain-specific, biased to maximally exploit compute resources, and not guaranteed to be reproducible. In this paper, we present a new framework of competition design that promotes the development of algorithms that overcome these barriers. We propose four central mechanisms for achieving this end: submission retraining, domain randomization, desemantization through domain obfuscation, and the limitation of competition compute and environment-sample budget. To demonstrate the efficacy of this design, we proposed, organized, and ran the MineRL 2020 Competition on Sample-Efficient Reinforcement Learning. In this work, we describe the organizational outcomes of the competition and show that the resulting participant submissions are reproducible, non-specific to the competition environment, and sample/resource efficient, despite the difficult competition task.

Current work in explainable reinforcement learning generally produces policies in the form of a decision tree over the state space. Such policies can be used for formal safety verification, agent behavior prediction, and manual inspection of important features. However, existing approaches fit a decision tree after training or use a custom learning procedure which is not compatible with new learning techniques, such as those which use neural networks. To address this limitation, we propose a novel Markov Decision Process (MDP) type for learning decision tree policies: Iterative Bounding MDPs (IBMDPs). An IBMDP is constructed around a base MDP so each IBMDP policy is guaranteed to correspond to a decision tree policy for the base MDP when using a method-agnostic masking procedure. Because of this decision tree equivalence, any function approximator can be used during training, including a neural network, while yielding a decision tree policy for the base MDP. We present the required masking procedure as well as a modified value update step which allows IBMDPs to be solved using existing algorithms. We apply this procedure to produce IBMDP variants of recent reinforcement learning methods. We empirically show the benefits of our approach by solving IBMDPs to produce decision tree policies for the base MDPs.

Although deep reinforcement learning has led to breakthroughs in many difficult domains, these successes have required an ever-increasing number of samples, affording only a shrinking segment of the AI community access to their development. Resolution of these limitations requires new, sample-efficient methods. To facilitate research in this direction, we propose this second iteration of the MineRL Competition. The primary goal of the competition is to foster the development of algorithms which can efficiently leverage human demonstrations to drastically reduce the number of samples needed to solve complex, hierarchical, and sparse environments. To that end, participants compete under a limited environment sample-complexity budget to develop systems which solve the MineRL ObtainDiamond task in Minecraft, a sequential decision making environment requiring long-term planning, hierarchical control, and efficient exploration methods. The competition is structured into two rounds in which competitors are provided several paired versions of the dataset and environment with different game textures and shaders. At the end of each round, competitors submit containerized versions of their learning algorithms to the AIcrowd platform where they are trained from scratch on a hold-out dataset-environment pair for a total of 4-days on a pre-specified hardware platform. In this follow-up iteration to the NeurIPS 2019 MineRL Competition, we implement new features to expand the scale and reach of the competition. In response to the feedback of the previous participants, we introduce a second minor track focusing on solutions without access to environment interactions of any kind except during test-time. Further we aim to prompt domain agnostic submissions by implementing several novel competition mechanics including action-space randomization and desemantization of observations and actions.

Democratizing access to artificial intelligence (AI) requires competitions that promote the development of sample-efficient learning, as well as ensure the reproducibility and generalizability of results. Sample efficiency is important because practitioners with limited compute resources cannot readily utilize algorithms that require a massive number of samples. The complexity of these stateof-the-art methods is outpacing advancements in computation. Moreover, as methods and domains become more specialized, learning procedures become more fragile: often undocumented modifications can inhibit reproducible results and seeds are chosen to reflect the optimal performance of a given solution [Henderson et al., 2018]. Because the focus of traditional research challenges is the development of new techniques in a particular field, these challenges seek to reward participants for novel solutions. However, submissions with the best performance on the (often highly specified) task tend leverage domain knowledge that is not broadly applicable, leading challenges to open separate tracks where submissions are subjectively evaluated on research novelty [Pavlov et al., 2018]. To encourage participants to develop methods with reproducible and robust training behavior, we propose a challenge paradigm where competitors are evaluated directly on the performance of their learning procedures rather than pre-trained agents. Since competition organizers retrain submissions in a controlled setting they can guarantee reproducibility, and - by retraining submissions using a held-out test set - help ensure generalization of submissions past the environments on which they were trained.

The sample inefficiency of standard deep reinforcement learning methods precludes their application to many real-world problems. Methods which leverage human demonstrations require fewer samples but have been researched less. As demonstrated in the computer vision and natural language processing communities, large-scale datasets have the capacity to facilitate research by serving as an experimental and benchmarking platform for new methods. However, existing datasets compatible with reinforcement learning simulators do not have sufficient scale, structure, and quality to enable the further development and evaluation of methods focused on using human examples. Therefore, we introduce a comprehensive, large-scale, simulator-paired dataset of human demonstrations: MineRL. The dataset consists of over 60 million automatically annotated state-action pairs across a variety of related tasks in Minecraft, a dynamic, 3D, open-world environment. We present a novel data collection scheme which allows for the ongoing introduction of new tasks and the gathering of complete state information suitable for a variety of methods. We demonstrate the hierarchality, diversity, and scale of the MineRL dataset. Further, we show the difficulty of the Minecraft domain along with the potential of MineRL in developing techniques to solve key research challenges within it.

There is a growing desire in the field of reinforcement learning (and machine learning in general) to move from black-box models toward more "interpretable AI." We improve interpretability of reinforcement learning by increasing the utility of decision tree policies learned via reinforcement learning. These policies consist of a decision tree over the state space, which requires fewer parameters to express than traditional policy representations. Existing methods for creating decision tree policies via reinforcement learning focus on accurately representing an action-value function during training, but this leads to much larger trees than would otherwise be required. To address this shortcoming, we propose a novel algorithm which only increases tree size when the estimated discounted future reward of the overall policy would increase by a sufficient amount. Through evaluation in a simulated environment, we show that its performance is comparable or superior to traditional tree-based approaches and that it yields a more succinct policy. Additionally, we discuss tuning parameters to control the tradeoff between optimizing for smaller tree size or for overall reward.

Though reinforcement learning has greatly benefited from the incorporation of neural networks, the inability to verify the correctness of such systems limits their use. Current work in explainable deep learning focuses on explaining only a single decision in terms of input features, making it unsuitable for explaining a sequence of decisions. To address this need, we introduce Abstracted Policy Graphs, which are Markov chains of abstract states. This representation concisely summarizes a policy so that individual decisions can be explained in the context of expected future transitions. Additionally, we propose a method to generate these Abstracted Policy Graphs for deterministic policies given a learned value function and a set of observed transitions, potentially off-policy transitions used during training. Since no restrictions are placed on how the value function is generated, our method is compatible with many existing reinforcement learning methods. We prove that the worst-case time complexity of our method is quadratic in the number of features and linear in the number of provided transitions, $O(|F|^2 |tr\_samples|)$. By applying our method to a family of domains, we show that our method scales well in practice and produces Abstracted Policy Graphs which reliably capture relationships within these domains.

Though deep reinforcement learning has led to breakthroughs in many difficult domains, these successes have required an ever-increasing number of samples. As state-of-the-art reinforcement learning (RL) systems require an exponentially increasing number of samples, their development is restricted to a continually shrinking segment of the AI community. Likewise, many of these systems cannot be applied to real-world problems, where environment samples are expensive. Resolution of these limitations requires new, sample-efficient methods. To facilitate research in this direction, we introduce the MineRL Competition on Sample Efficient Reinforcement Learning using Human Priors. The primary goal of the competition is to foster the development of algorithms which can efficiently leverage human demonstrations to drastically reduce the number of samples needed to solve complex, hierarchical, and sparse environments. To that end, we introduce: (1) the Minecraft ObtainDiamond task, a sequential decision making environment requiring long-term planning, hierarchical control, and efficient exploration methods; and (2) the MineRL-v0 dataset, a large-scale collection of over 60 million state-action pairs of human demonstrations that can be resimulated into embodied trajectories with arbitrary modifications to game state and visuals. Participants will compete to develop systems which solve the ObtainDiamond task with a limited number of samples from the environment simulator, Malmo. The competition is structured into two rounds in which competitors are provided several paired versions of the dataset and environment with different game textures. At the end of each round, competitors will submit containerized versions of their learning algorithms and they will then be trained/evaluated from scratch on a hold-out dataset-environment pair for a total of 4-days on a prespecified hardware platform.

In this paper, we show a phenomenon, which we named "super-convergence", where residual networks can be trained using an order of magnitude fewer iterations than is used with standard training methods. The existence of super-convergence is relevant to understanding why deep networks generalize well. One of the key elements of super-convergence is training with cyclical learning rates and a large maximum learning rate. Furthermore, we present evidence that training with large learning rates improves performance by regularizing the network. In addition, we show that super-convergence provides a greater boost in performance relative to standard training when the amount of labeled training data is limited. We also derive a simplification of the Hessian Free optimization method to compute an estimate of the optimal learning rate. The architectures and code to replicate the figures in this paper are available at github.com/lnsmith54/super-convergence.