The emergence of latency-critical AI applications has been supported by the evolution of the edge computing paradigm. However, edge solutions are typically resource-constrained, posing reliability challenges due to heightened contention for compute and communication capacities and faulty application behavior in the presence of overload conditions. Although a large amount of generated log data can be mined for fault prediction, labeling this data for training is a manual process and thus a limiting factor for automation. Due to this, many companies resort to unsupervised fault-tolerance models. Yet, failure models of this kind can incur a loss of accuracy when they need to adapt to non-stationary workloads and diverse host characteristics. To cope with this, we propose a novel modeling approach, called DeepFT, to proactively avoid system overloads and their adverse effects by optimizing the task scheduling and migration decisions. DeepFT uses a deep surrogate model to accurately predict and diagnose faults in the system and co-simulation based self-supervised learning to dynamically adapt the model in volatile settings. It offers a highly scalable solution as the model size scales by only 3 and 1 percent per unit increase in the number of active tasks and hosts. Extensive experimentation on a Raspberry-Pi based edge cluster with DeFog benchmarks shows that DeepFT can outperform state-of-the-art baseline methods in fault-detection and QoS metrics. Specifically, DeepFT gives the highest F1 scores for fault-detection, reducing service deadline violations by up to 37\% while also improving response time by up to 9%.

Edge Federation是一种新的计算范式，无缝地互连多个边缘服务提供商的资源。此类系统中的一个关键挑战是在受约束设备中部署基于延迟和AI的资源密集型应用程序。为了应对这一挑战，我们提出了一种新型的基于记忆有效的深度学习模型，即生成优化网络（GON）。与甘斯不同，成人使用单个网络既区分输入又生成样本，从而大大降低了它们的内存足迹。利用奇数的低内存足迹，我们提出了一种称为Dragon的分散性故障耐受性方法，该方法运行模拟（按照数字建模双胞胎）来快速预测和优化边缘联邦的性能。在多个基于Raspberry-Pi的联合边缘配置上使用现实世界边缘计算基准测试的广泛实验表明，龙可以胜过故障检测和服务质量（QOS）指标的基线方法。具体而言，所提出的方法给出了与最佳深度学习方法（DL）方法更高的F1分数，而与启发式方法相比，记忆力较低。这使得违反能源消耗，响应时间和服务水平协议分别提高了74％，63％和82％。

最近，已经提出了使用代理模型的智能调度方法，以便在异构雾环境中有效地分配易失性任务。确定性代理模型，深神经网络（DNN）和基于梯度的优化等进步允许达到低能量消耗和响应时间。然而，确定估计优化的客观值的确定性代理模型，不考虑可以导致高服务级别协议（SLA）违规率的服务质量（QoS）目标函数的不确定性。此外，DNN训练的脆性性质，防止这些模型达到最小的能量或响应时间。为了克服这些困难，我们提出了一种新的调度程序：GOSH I.E.使用二阶衍生物和异源塑料深层代理模型的梯度优化。 GOSH使用二阶梯度基于基于梯度的优化方法来获得更好的QoS并减少迭代的次数，以收敛到调度决定，随后降低调度时间。 GOSH而不是Vanilla DNN，使用自然参数网络来近似客观分数。此外，较低的置信度优化方法可以通过采用基于误差的探索来在贪婪最小化和不确定性降低之间找到最佳权衡。因此，GOSH及其共模的扩展GOSH *可以快速调整并达到比基线方法更好的客观评分。我们表明GOSH *达到比GOSH更好的客观分数，但它仅适用于高资源可用性设置，而GOSH则适用于有限的资源设置。 GOSH和GOSH的真实系统实验*在能源消耗，响应时间和SLA分别违反最多18,27和82％的情况下，对最先进的技术进行了显着改善。

工作流程调度是一个并行和分布式计算（PDC）的长期研究，旨在有效地利用计算资源来满足用户的服务要求。最近提出的调度方法利用边缘计算平台的低响应时间来优化服务质量（QoS）。然而，由于计算异质性，移动设备的延迟以及工作负载资源要求的挥发性，因此由于计算异质性而挑战，在移动边缘云系统中的调度工作流程应用是具有挑战性的。为了克服这些困难，它是必不可少的，但同时具有挑战性，开发一种有效地模拟QoS目标的长视力优化方案。在这项工作中，我们提出了MCDS：Monte Carlo学习使用Deep代理模型来有效地安排移动边缘云计算系统中的工作流程应用。 MCD是一种基于人工智能（AI）的调度方法，它使用基于树的搜索策略和基于深度神经网络的代理模型来估计即时动作的长期QoS影响，以实现调度决策的鲁棒优化。物理和模拟边缘云试验台的实验表明，MCD在能耗，响应时间，SLA违规方面可以改善最先进的方法，违规和成本分别至少为6.13,4.56,45.09和30.71％。

由于边缘设备的不可靠性以及现代应用的严格的服务截止日期，构建一个容错的边缘系统可以快速地对节点过载或故障发生的挑战是具有挑战性的。此外，不必要的任务迁移可能会强调系统网络，从而强调需要智能和解析故障恢复方案。现有方法通常无法适应高度挥发性的工作量或准确地检测和诊断故障以获得最佳修复。因此，需要一种坚固且主动的容错机制来满足服务级别目标。在这项工作中，我们提出了一种使用生成的对冲网络（GaN）的复合AI模型来预测集装箱边缘部署中的主动容错的抢占迁移决策。 Pregan使用串联的共同模拟与GaN一起学习几次异常的分类器，并主动预测可靠计算的迁移决策。基于Raspberry-PI的边缘环境的广泛实验表明，Pregan可以在故障检测，诊断和分类中优于最先进的基线方法，从而实现高质量的服务。与所考虑的基线中的最佳方法相比，Pregan完成了5.1％的准确故障检测，更高的诊断得分和23.8％的开销。

The highest grossing media franchise of all times, with over \$90 billion in total revenue, is Pokemon. The video games belong to the class of Japanese Role Playing Games (J-RPG). Developing a powerful AI agent for these games is very hard because they present big challenges to MinMax, Monte Carlo Tree Search and statistical Machine Learning, as they are vastly different from the well explored in AI literature games. An AI agent for one of these games means significant progress in AI agents for the entire class. Further, the key principles of such work can hopefully inspire approaches to several domains that require excellent teamwork under conditions of extreme uncertainty, including managing a team of doctors, robots or employees in an ever changing environment, like a pandemic stricken region or a war-zone. In this paper we first explain the mechanics of the game and we perform a game analysis. We continue by proposing unique AI algorithms based on our understanding that the two biggest challenges in the game are keeping a balanced team and dealing with three sources of uncertainty. Later on, we describe why evaluating the performance of such agents is challenging and we present the results of our approach. Our AI agent performed significantly better than all previous attempts and peaked at the 33rd place in the world, in one of the most popular battle formats, while running on only 4 single socket servers.

As language models (LMs) scale, they develop many novel behaviors, good and bad, exacerbating the need to evaluate how they behave. Prior work creates evaluations with crowdwork (which is time-consuming and expensive) or existing data sources (which are not always available). Here, we automatically generate evaluations with LMs. We explore approaches with varying amounts of human effort, from instructing LMs to write yes/no questions to making complex Winogender schemas with multiple stages of LM-based generation and filtering. Crowdworkers rate the examples as highly relevant and agree with 90-100% of labels, sometimes more so than corresponding human-written datasets. We generate 154 datasets and discover new cases of inverse scaling where LMs get worse with size. Larger LMs repeat back a dialog user's preferred answer ("sycophancy") and express greater desire to pursue concerning goals like resource acquisition and goal preservation. We also find some of the first examples of inverse scaling in RL from Human Feedback (RLHF), where more RLHF makes LMs worse. For example, RLHF makes LMs express stronger political views (on gun rights and immigration) and a greater desire to avoid shut down. Overall, LM-written evaluations are high-quality and let us quickly discover many novel LM behaviors.

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

As AI systems become more capable, we would like to enlist their help to supervise other AIs. We experiment with methods for training a harmless AI assistant through self-improvement, without any human labels identifying harmful outputs. The only human oversight is provided through a list of rules or principles, and so we refer to the method as 'Constitutional AI'. The process involves both a supervised learning and a reinforcement learning phase. In the supervised phase we sample from an initial model, then generate self-critiques and revisions, and then finetune the original model on revised responses. In the RL phase, we sample from the finetuned model, use a model to evaluate which of the two samples is better, and then train a preference model from this dataset of AI preferences. We then train with RL using the preference model as the reward signal, i.e. we use 'RL from AI Feedback' (RLAIF). As a result we are able to train a harmless but non-evasive AI assistant that engages with harmful queries by explaining its objections to them. Both the SL and RL methods can leverage chain-of-thought style reasoning to improve the human-judged performance and transparency of AI decision making. These methods make it possible to control AI behavior more precisely and with far fewer human labels.

Developing safe and useful general-purpose AI systems will require us to make progress on scalable oversight: the problem of supervising systems that potentially outperform us on most skills relevant to the task at hand. Empirical work on this problem is not straightforward, since we do not yet have systems that broadly exceed our abilities. This paper discusses one of the major ways we think about this problem, with a focus on how to turn it into one that can be productively studied empirically. We first present an experimental design centered on choosing tasks for which human specialists succeed but unaided humans and current general AI systems fail. We then present a proof-of-concept experiment following meant to demonstrate a key feature of this experimental design and show its viability with two question-answering tasks: MMLU and time-limited QuALITY. On these tasks, we find that human participants who interact with an unreliable large-language-model dialog assistant through chat -- a trivial baseline strategy for scalable oversight -- substantially outperform both the model alone and their own unaided performance. These results are an encouraging sign that scalable oversight will be tractable to study with present models and bolster recent findings that large language models can productively assist humans with difficult tasks.