允许代理商通过沟通共享信息对于解决多代理增强学习中的复杂任务至关重要。在这项工作中，我们考虑了给定通信协议是否可以表达任意政策的问题。通过观察许多现有协议可以看作是图神经网络（GNN）的实例，我们证明了联合动作选择与节点标记的等效性。通过证明其表达能力的标准GNN方法，我们从现有的GNN文献中汲取了限制，并考虑使用以下方式增强剂观察：（1）独特的代理ID和（2）随机噪声。我们提供了有关这些方法如何产生普遍表达性交流的理论分析，并证明它们能够针对相同代理的任意行动集。从经验上讲，这些增强被发现可以改善需要表达性交流的任务的性能，而通常发现最佳通信协议是任务依赖性的。

在各种现实世界应用中，组合优化问题作为混合整数线性程序（MILP）无处不在。规范的分支和结合算法通过构建越来越约束的子问题的搜索树来寻求精确解决MILP。实际上，其解决时间性能取决于启发式方法，例如选择下一个变量来约束（“分支”）。最近，机器学习（ML）已成为分支的有希望的范式。但是，先前的工作一直在努力应用强化学习（RL），理由是稀疏的奖励，艰难的探索和部分可观察性是重大挑战。取而代之的是，领先的ML方法论通过模仿学习（IL）近似高质量的手工启发式方法，这排除了新型政策的发现并需要昂贵的数据标签。在这项工作中，我们提出了复古分支。一种简单而有效的分支RL方法。通过回顾性将搜索树解构为子树中包含的多个路径，我们使代理能够从更短的轨迹中学习具有更可预测的下一步状态。在对四个组合任务的实验中，我们的方法可以在没有任何专家指导或预培训的情况下学习分支。我们的表现优于当前最新的RL分支算法，比最佳IL方法在MILPS上具有500个约束和1000个变量的最佳性能的20％以内，并验证了我们的回顾性构建轨迹对于实现的必要这些结果。

This project leverages advances in multi-agent reinforcement learning (MARL) to improve the efficiency and flexibility of order-picking systems for commercial warehouses. We envision a warehouse of the future in which dozens of mobile robots and human pickers work together to collect and deliver items within the warehouse. The fundamental problem we tackle, called the order-picking problem, is how these worker agents must coordinate their movement and actions in the warehouse to maximise performance (e.g. order throughput) under given resource constraints. Established industry methods using heuristic approaches require large engineering efforts to optimise for innately variable warehouse configurations. In contrast, the MARL framework can be flexibly applied to any warehouse configuration (e.g. size, layout, number/types of workers, item replenishment frequency) and the agents learn via a process of trial-and-error how to optimally cooperate with one another. This paper details the current status of the R&D effort initiated by Dematic and the University of Edinburgh towards a general-purpose and scalable MARL solution for the order-picking problem in realistic warehouses.

We study representation learning for efficient imitation learning over linear systems. In particular, we consider a setting where learning is split into two phases: (a) a pre-training step where a shared $k$-dimensional representation is learned from $H$ source policies, and (b) a target policy fine-tuning step where the learned representation is used to parameterize the policy class. We find that the imitation gap over trajectories generated by the learned target policy is bounded by $\tilde{O}\left( \frac{k n_x}{HN_{\mathrm{shared}}} + \frac{k n_u}{N_{\mathrm{target}}}\right)$, where $n_x > k$ is the state dimension, $n_u$ is the input dimension, $N_{\mathrm{shared}}$ denotes the total amount of data collected for each policy during representation learning, and $N_{\mathrm{target}}$ is the amount of target task data. This result formalizes the intuition that aggregating data across related tasks to learn a representation can significantly improve the sample efficiency of learning a target task. The trends suggested by this bound are corroborated in simulation.

Purpose: Trans-oral robotic surgery (TORS) using the da Vinci surgical robot is a new minimally-invasive surgery method to treat oropharyngeal tumors, but it is a challenging operation. Augmented reality (AR) based on intra-operative ultrasound (US) has the potential to enhance the visualization of the anatomy and cancerous tumors to provide additional tools for decision-making in surgery. Methods: We propose and carry out preliminary evaluations of a US-guided AR system for TORS, with the transducer placed on the neck for a transcervical view. Firstly, we perform a novel MRI-transcervical 3D US registration study. Secondly, we develop a US-robot calibration method with an optical tracker and an AR system to display the anatomy mesh model in the real-time endoscope images inside the surgeon console. Results: Our AR system reaches a mean projection error of 26.81 and 27.85 pixels for the projection from the US to stereo cameras in a water bath experiment. The average target registration error for MRI to 3D US is 8.90 mm for the 3D US transducer and 5.85 mm for freehand 3D US, and the average distance between the vessel centerlines is 2.32 mm. Conclusion: We demonstrate the first proof-of-concept transcervical US-guided AR system for TORS and the feasibility of trans-cervical 3D US-MRI registration. Our results show that trans-cervical 3D US is a promising technique for TORS image guidance.

Reservoir computing is a recurrent neural network paradigm in which only the output layer is trained. Recently, it was demonstrated that adding time-shifts to the signals generated by a reservoir can provide large improvements in performance accuracy. In this work, we present a technique to choose the optimal time shifts. Our technique maximizes the rank of the reservoir matrix using a rank-revealing QR algorithm and is not task dependent. Further, our technique does not require a model of the system, and therefore is directly applicable to analog hardware reservoir computers. We demonstrate our time-shift optimization technique on two types of reservoir computer: one based on an opto-electronic oscillator and the traditional recurrent network with a $tanh$ activation function. We find that our technique provides improved accuracy over random time-shift selection in essentially all cases.

Network intrusion detection systems (NIDS) to detect malicious attacks continues to meet challenges. NIDS are vulnerable to auto-generated port scan infiltration attempts and NIDS are often developed offline, resulting in a time lag to prevent the spread of infiltration to other parts of a network. To address these challenges, we use hypergraphs to capture evolving patterns of port scan attacks via the set of internet protocol addresses and destination ports, thereby deriving a set of hypergraph-based metrics to train a robust and resilient ensemble machine learning (ML) NIDS that effectively monitors and detects port scanning activities and adversarial intrusions while evolving intelligently in real-time. Through the combination of (1) intrusion examples, (2) NIDS update rules, (3) attack threshold choices to trigger NIDS retraining requests, and (4) production environment with no prior knowledge of the nature of network traffic 40 scenarios were auto-generated to evaluate the ML ensemble NIDS comprising three tree-based models. Results show that under the model settings of an Update-ALL-NIDS rule (namely, retrain and update all the three models upon the same NIDS retraining request) the proposed ML ensemble NIDS produced the best results with nearly 100% detection performance throughout the simulation, exhibiting robustness in the complex dynamics of the simulated cyber-security scenario.

Human behavior emerges from planning over elaborate decompositions of tasks into goals, subgoals, and low-level actions. How are these decompositions created and used? Here, we propose and evaluate a normative framework for task decomposition based on the simple idea that people decompose tasks to reduce the overall cost of planning while maintaining task performance. Analyzing 11,117 distinct graph-structured planning tasks, we find that our framework justifies several existing heuristics for task decomposition and makes predictions that can be distinguished from two alternative normative accounts. We report a behavioral study of task decomposition ($N=806$) that uses 30 randomly sampled graphs, a larger and more diverse set than that of any previous behavioral study on this topic. We find that human responses are more consistent with our framework for task decomposition than alternative normative accounts and are most consistent with a heuristic -- betweenness centrality -- that is justified by our approach. Taken together, our results provide new theoretical insight into the computational principles underlying the intelligent structuring of goal-directed behavior.

机器学习潜力是分子模拟的重要工具，但是由于缺乏高质量数据集来训练它们的发展，它们的开发阻碍了它们。我们描述了Spice数据集，这是一种新的量子化学数据集，用于训练与模拟与蛋白质相互作用的药物样的小分子相关的潜在。它包含超过110万个小分子，二聚体，二肽和溶剂化氨基酸的构象。它包括15个元素，带电和未充电的分子以及广泛的共价和非共价相互作用。它提供了在{\ omega} b97m-d3（bj）/def2-tzVPPD理论水平以及其他有用的数量（例如多极矩和键阶）上计算出的力和能量。我们在其上训练一组机器学习潜力，并证明它们可以在化学空间的广泛区域中实现化学精度。它可以作为创建可转移的，准备使用潜在功能用于分子模拟的宝贵资源。

基于注意力的神经网络在许多AI任务中都普遍存在。尽管其出色的算法性能，但注意力机制和前馈网络（FFN）的使用仍需要过多的计算和内存资源，这通常会损害其硬件性能。尽管已经引入了各种稀疏变体，但大多数方法仅着重于缓解算法级别上的二次注意力缩放，而无需明确考虑将其方法映射到真实硬件设计上的效率。此外，大多数努力仅专注于注意机制或FFN，但没有共同优化这两个部分，导致当前的大多数设计在处理不同的输入长度时缺乏可扩展性。本文从硬件角度系统地考虑了不同变体中的稀疏模式。在算法级别上，我们提出了Fabnet，这是一种适合硬件的变体，它采用统一的蝴蝶稀疏模式来近似关注机制和FFN。在硬件级别上，提出了一种新颖的适应性蝴蝶加速器，可以在运行时通过专用硬件控件配置，以使用单个统一的硬件引擎加速不同的蝴蝶层。在远程 - ARENA数据集上，FabNet达到了与香草变压器相同的精度，同时将计算量减少10到66次，参数数量为2至22次。通过共同优化算法和硬件，我们的基于FPGA的蝴蝶加速器在归一化到同一计算预算的最新加速器上达到了14.2至23.2倍的速度。与Raspberry Pi 4和Jetson Nano上优化的CPU和GPU设计相比，我们的系统在相同的功率预算下的最大273.8和15.1倍。