DeepMind的游戏理论与多代理团队研究多学科学习的几个方面，从计算近似值到游戏理论中的基本概念，再到在富裕的空间环境中模拟社会困境，并在困难的团队协调任务中培训3-D类人动物。我们小组的一个签名目的是使用DeepMind在DeepMind中提供的资源和专业知识，以深入强化学习来探索复杂环境中的多代理系统，并使用这些基准来提高我们的理解。在这里，我们总结了我们团队的最新工作，并提出了一种分类法，我们认为这重点介绍了多代理研究中许多重要的开放挑战。

当今许多大型系统的设计，从交通路由环境到智能电网，都依赖游戏理论平衡概念。但是，随着$ n $玩家游戏的大小通常会随着$ n $而成倍增长，标准游戏理论分析实际上是不可行的。最近的方法通过考虑平均场游戏，匿名$ n $玩家游戏的近似值，在这种限制中，玩家的数量是无限的，而人口的状态分布，而不是每个单独的球员的状态，是兴趣。然而，迄今为止研究最多的平均场平衡的平均场nash平衡的实际可计算性通常取决于有益的非一般结构特性，例如单调性或收缩性能，这是已知的算法收敛所必需的。在这项工作中，我们通过开发均值相关和与粗相关的平衡的概念来研究平均场比赛的替代途径。我们证明，可以使用三种经典算法在\ emph {ash All Games}中有效地学习它们，而无需对游戏结构进行任何其他假设。此外，我们在文献中已经建立了对应关系，从而获得了平均场 - $ n $玩家过渡的最佳范围，并经验证明了这些算法在简单游戏中的收敛性。

我们介绍了DeepNash，这是一种能够学习从头开始播放不完美的信息游戏策略的自主代理，直到人类的专家级别。 Stratego是人工智能（AI）尚未掌握的少数标志性棋盘游戏之一。这个受欢迎的游戏具有$ 10^{535} $节点的巨大游戏树，即，$ 10^{175} $倍的$倍于GO。它具有在不完美的信息下需要决策的其他复杂性，类似于德克萨斯州Hold'em扑克，该扑克的游戏树较小（以$ 10^{164} $节点为单位）。 Stratego中的决策是在许多离散的动作上做出的，而动作与结果之间没有明显的联系。情节很长，在球员获胜之前经常有数百次动作，而Stratego中的情况则不能像扑克中那样轻松地分解成管理大小的子问题。由于这些原因，Stratego几十年来一直是AI领域的巨大挑战，现有的AI方法几乎没有达到业余比赛水平。 Deepnash使用游戏理论，无模型的深钢筋学习方法，而无需搜索，该方法学会通过自我播放来掌握Stratego。 DeepNash的关键组成部分的正则化NASH Dynamics（R-NAD）算法通过直接修改基础多项式学习动力学来收敛到近似NASH平衡，而不是围绕它“循环”。 Deepnash在Stratego中击败了现有的最先进的AI方法，并在Gravon Games平台上获得了年度（2022年）和历史前3名，并与人类专家竞争。

Markowitz mean-variance portfolios with sample mean and covariance as input parameters feature numerous issues in practice. They perform poorly out of sample due to estimation error, they experience extreme weights together with high sensitivity to change in input parameters. The heavy-tail characteristics of financial time series are in fact the cause for these erratic fluctuations of weights that consequently create substantial transaction costs. In robustifying the weights we present a toolbox for stabilizing costs and weights for global minimum Markowitz portfolios. Utilizing a projected gradient descent (PGD) technique, we avoid the estimation and inversion of the covariance operator as a whole and concentrate on robust estimation of the gradient descent increment. Using modern tools of robust statistics we construct a computationally efficient estimator with almost Gaussian properties based on median-of-means uniformly over weights. This robustified Markowitz approach is confirmed by empirical studies on equity markets. We demonstrate that robustified portfolios reach the lowest turnover compared to shrinkage-based and constrained portfolios while preserving or slightly improving out-of-sample performance.

Recommendation Systems (RSs) are ubiquitous in modern society and are one of the largest points of interaction between humans and AI. Modern RSs are often implemented using deep learning models, which are infamously difficult to interpret. This problem is particularly exasperated in the context of recommendation scenarios, as it erodes the user's trust in the RS. In contrast, the newly introduced Tsetlin Machines (TM) possess some valuable properties due to their inherent interpretability. TMs are still fairly young as a technology. As no RS has been developed for TMs before, it has become necessary to perform some preliminary research regarding the practicality of such a system. In this paper, we develop the first RS based on TMs to evaluate its practicality in this application domain. This paper compares the viability of TMs with other machine learning models prevalent in the field of RS. We train and investigate the performance of the TM compared with a vanilla feed-forward deep learning model. These comparisons are based on model performance, interpretability/explainability, and scalability. Further, we provide some benchmark performance comparisons to similar machine learning solutions relevant to RSs.

We propose an approach for semantic imitation, which uses demonstrations from a source domain, e.g. human videos, to accelerate reinforcement learning (RL) in a different target domain, e.g. a robotic manipulator in a simulated kitchen. Instead of imitating low-level actions like joint velocities, our approach imitates the sequence of demonstrated semantic skills like "opening the microwave" or "turning on the stove". This allows us to transfer demonstrations across environments (e.g. real-world to simulated kitchen) and agent embodiments (e.g. bimanual human demonstration to robotic arm). We evaluate on three challenging cross-domain learning problems and match the performance of demonstration-accelerated RL approaches that require in-domain demonstrations. In a simulated kitchen environment, our approach learns long-horizon robot manipulation tasks, using less than 3 minutes of human video demonstrations from a real-world kitchen. This enables scaling robot learning via the reuse of demonstrations, e.g. collected as human videos, for learning in any number of target domains.

By transferring knowledge from large, diverse, task-agnostic datasets, modern machine learning models can solve specific downstream tasks either zero-shot or with small task-specific datasets to a high level of performance. While this capability has been demonstrated in other fields such as computer vision, natural language processing or speech recognition, it remains to be shown in robotics, where the generalization capabilities of the models are particularly critical due to the difficulty of collecting real-world robotic data. We argue that one of the keys to the success of such general robotic models lies with open-ended task-agnostic training, combined with high-capacity architectures that can absorb all of the diverse, robotic data. In this paper, we present a model class, dubbed Robotics Transformer, that exhibits promising scalable model properties. We verify our conclusions in a study of different model classes and their ability to generalize as a function of the data size, model size, and data diversity based on a large-scale data collection on real robots performing real-world tasks. The project's website and videos can be found at robotics-transformer.github.io

Foveated imaging provides a better tradeoff between situational awareness (field of view) and resolution and is critical in long-wavelength infrared regimes because of the size, weight, power, and cost of thermal sensors. We demonstrate computational foveated imaging by exploiting the ability of a meta-optical frontend to discriminate between different polarization states and a computational backend to reconstruct the captured image/video. The frontend is a three-element optic: the first element which we call the "foveal" element is a metalens that focuses s-polarized light at a distance of $f_1$ without affecting the p-polarized light; the second element which we call the "perifoveal" element is another metalens that focuses p-polarized light at a distance of $f_2$ without affecting the s-polarized light. The third element is a freely rotating polarizer that dynamically changes the mixing ratios between the two polarization states. Both the foveal element (focal length = 150mm; diameter = 75mm), and the perifoveal element (focal length = 25mm; diameter = 25mm) were fabricated as polarization-sensitive, all-silicon, meta surfaces resulting in a large-aperture, 1:6 foveal expansion, thermal imaging capability. A computational backend then utilizes a deep image prior to separate the resultant multiplexed image or video into a foveated image consisting of a high-resolution center and a lower-resolution large field of view context. We build a first-of-its-kind prototype system and demonstrate 12 frames per second real-time, thermal, foveated image, and video capture in the wild.

Large-scale data is an essential component of machine learning as demonstrated in recent advances in natural language processing and computer vision research. However, collecting large-scale robotic data is much more expensive and slower as each operator can control only a single robot at a time. To make this costly data collection process efficient and scalable, we propose Policy Assisted TeleOperation (PATO), a system which automates part of the demonstration collection process using a learned assistive policy. PATO autonomously executes repetitive behaviors in data collection and asks for human input only when it is uncertain about which subtask or behavior to execute. We conduct teleoperation user studies both with a real robot and a simulated robot fleet and demonstrate that our assisted teleoperation system reduces human operators' mental load while improving data collection efficiency. Further, it enables a single operator to control multiple robots in parallel, which is a first step towards scalable robotic data collection. For code and video results, see https://clvrai.com/pato

Class-Incremental Learning is a challenging problem in machine learning that aims to extend previously trained neural networks with new classes. This is especially useful if the system is able to classify new objects despite the original training data being unavailable. While the semantic segmentation problem has received less attention than classification, it poses distinct problems and challenges since previous and future target classes can be unlabeled in the images of a single increment. In this case, the background, past and future classes are correlated and there exist a background-shift. In this paper, we address the problem of how to model unlabeled classes while avoiding spurious feature clustering of future uncorrelated classes. We propose to use Evidential Deep Learning to model the evidence of the classes as a Dirichlet distribution. Our method factorizes the problem into a separate foreground class probability, calculated by the expected value of the Dirichlet distribution, and an unknown class (background) probability corresponding to the uncertainty of the estimate. In our novel formulation, the background probability is implicitly modeled, avoiding the feature space clustering that comes from forcing the model to output a high background score for pixels that are not labeled as objects. Experiments on the incremental Pascal VOC, and ADE20k benchmarks show that our method is superior to state-of-the-art, especially when repeatedly learning new classes with increasing number of increments.