简介:血管可以从数字眼底图像(DFI)中可视化。几项研究表明,从DFI获得的心血管风险与血管特征之间存在关联。计算机视觉和图像分割的最新进展使自动化DFI血管分割。需要从这些分段DFI中自动计算数字脉管生物标志物(VBM)的资源。方法:在本文中,我们引入了Python Vasculature生物标志物工具箱,表示为PVBM。总共实施了11个VBM。特别是,我们引入了新的算法方法来估计曲折和分支角度。使用PVBM和作为可用性的证明,我们分析了青光眼患者和健康对照组之间的几何血管差异。结果:我们基于DFI分割构建了一个全自动的血管生物标志物工具箱,并提供了表征青光眼的血管变化的可用性证明。对于小动脉和静脉,与健康对照组相比,青光眼患者的所有生物标志物都显着且较低,除了曲折度,静脉奇异长度和静脉分支角度。结论:我们已经从视网膜血管分割中对11个VBM进行了自动化。 PVBM工具箱是根据GNU GPL 3许可证的开源,可在Physiozoo.com(发布之后)上找到。
translated by 谷歌翻译
生成对抗性示例是创建添加到分类神经网络的输入信号的噪声的领域,从而改变网络的分类,同时保持噪声尽可能脆弱。虽然该主题在2D制度中得到了很好的研究,但它在3D制度中滞后,即攻击适用于3D点云或网格的分类网络,例如,对人们的3D扫描的姿势进行分类。截至目前,绝大多数论文都通过优化方法描述了这一制度的对抗性攻击。在本技术报告中,我们建议一个产生攻击的神经网络。该网络利用PointNet的体系结构进行一些更改。虽然我们基于我们的工作的前一篇文章必须单独优化每个形状,但没有任何学习的每个单独输入从头定制攻击,我们试图创建一个统一的模型,可以用一个向前推断所需的对抗性示例跑。
translated by 谷歌翻译
Function approximation has enabled remarkable advances in applying reinforcement learning (RL) techniques in environments with high-dimensional inputs, such as images, in an end-to-end fashion, mapping such inputs directly to low-level control. Nevertheless, these have proved vulnerable to small adversarial input perturbations. A number of approaches for improving or certifying robustness of end-to-end RL to adversarial perturbations have emerged as a result, focusing on cumulative reward. However, what is often at stake in adversarial scenarios is the violation of fundamental properties, such as safety, rather than the overall reward that combines safety with efficiency. Moreover, properties such as safety can only be defined with respect to true state, rather than the high-dimensional raw inputs to end-to-end policies. To disentangle nominal efficiency and adversarial safety, we situate RL in deterministic partially-observable Markov decision processes (POMDPs) with the goal of maximizing cumulative reward subject to safety constraints. We then propose a partially-supervised reinforcement learning (PSRL) framework that takes advantage of an additional assumption that the true state of the POMDP is known at training time. We present the first approach for certifying safety of PSRL policies under adversarial input perturbations, and two adversarial training approaches that make direct use of PSRL. Our experiments demonstrate both the efficacy of the proposed approach for certifying safety in adversarial environments, and the value of the PSRL framework coupled with adversarial training in improving certified safety while preserving high nominal reward and high-quality predictions of true state.
translated by 谷歌翻译
Generalist models, which are capable of performing diverse multi-modal tasks in a task-agnostic way within a single model, have been explored recently. Being, hopefully, an alternative to approaching general-purpose AI, existing generalist models are still at an early stage, where modality and task coverage is limited. To empower multi-modal task-scaling and speed up this line of research, we release a generalist model learning system, OFASys, built on top of a declarative task interface named multi-modal instruction. At the core of OFASys is the idea of decoupling multi-modal task representations from the underlying model implementations. In OFASys, a task involving multiple modalities can be defined declaratively even with just a single line of code. The system automatically generates task plans from such instructions for training and inference. It also facilitates multi-task training for diverse multi-modal workloads. As a starting point, we provide presets of 7 different modalities and 23 highly-diverse example tasks in OFASys, with which we also develop a first-in-kind, single model, OFA+, that can handle text, image, speech, video, and motion data. The single OFA+ model achieves 95% performance in average with only 16% parameters of 15 task-finetuned models, showcasing the performance reliability of multi-modal task-scaling provided by OFASys. Available at https://github.com/OFA-Sys/OFASys
translated by 谷歌翻译
Many problems can be viewed as forms of geospatial search aided by aerial imagery, with examples ranging from detecting poaching activity to human trafficking. We model this class of problems in a visual active search (VAS) framework, which takes as input an image of a broad area, and aims to identify as many examples of a target object as possible. It does this through a limited sequence of queries, each of which verifies whether an example is present in a given region. We propose a reinforcement learning approach for VAS that leverages a collection of fully annotated search tasks as training data to learn a search policy, and combines features of the input image with a natural representation of active search state. Additionally, we propose domain adaptation techniques to improve the policy at decision time when training data is not fully reflective of the test-time distribution of VAS tasks. Through extensive experiments on several satellite imagery datasets, we show that the proposed approach significantly outperforms several strong baselines. Code and data will be made public.
translated by 谷歌翻译
大多数强化学习算法隐含地假设强同步。我们提出了针对Q学习的新颖攻击,该攻击通过延迟有限时间段的奖励信号来利用该假设所带来的漏洞。我们考虑了两种类型的攻击目标:目标攻击,旨在使目标政策被学习,以及不靶向的攻击,这只是旨在诱使奖励低的政策。我们通过一系列实验评估了提出的攻击的功效。我们的第一个观察结果是,当目标仅仅是为了最大程度地减少奖励时,奖励延迟​​攻击非常有效。的确,我们发现即使是天真的基线奖励 - 延迟攻击也在最大程度地减少奖励方面也非常成功。另一方面,有针对性的攻击更具挑战性,尽管我们表明,提出的方法在实现攻击者的目标方面仍然非常有效。此外,我们引入了第二个威胁模型,该模型捕获了一种最小的缓解措施,该模型可确保不能超出顺序使用奖励。我们发现,这种缓解仍然不足以确保稳定性延迟但保留奖励的命令。
translated by 谷歌翻译
无源的无监督域适应性(SFUDA)旨在使用未标记的目标数据和训练有素的源域模型来学习目标域模型。大多数先前的SFUDA都致力于根据源知识推断目标数据的语义。在不衡量源知识的可传递性的情况下,这些方法不足以利用源知识,并且无法识别推断的目标语义的可靠性。但是,现有的可传递性测量需要源数据或目标标签,而SFUDA中是不可行的。为此,首先,我们提出了一种新颖的不确定性诱导的可传递性表示(UTR),该表示在没有源数据和目标标签的情况下,它利用不确定性作为工具来分析源编码的通道可传递性。域级UTR揭开了编码器通道向目标域的可传输程度,实例级别的UTR表征了推断的目标语义的可靠性。其次,基于UTR,我们为SFUDA提出了一个新颖的校准自适应框架(CAF),包括i)源知识校准模块,该模块指导目标模型学习可转移的源知识并丢弃不可转移的源知识,并且II)校准不可靠语义的目标语义校准模块。在校准的源知识和目标语义的帮助下,该模型可以安全地适应目标领域。我们使用实验结果验证了方法的有效性,并证明所提出的方法在三个SFUDA基准上实现了最先进的性能。代码可在https://github.com/spiresearch/utr上找到。
translated by 谷歌翻译
现有域适应方法假设域差异是由一些离散属性和变化引起的很少的离散属性。因此,我们建议研究一个新问题,即通过连续变化的属性形成无限结构域的晶状体连续域适应(CDA)。利用两个标记的源域和几个观察到的未标记目标域数据的知识,CDA的目的是学习具有连续属性的整个数据分布的通用模型。除了提出新问题的贡献外,我们还提出了一种新颖的方法作为强大的CDA基线。具体而言,首先,我们提出了一种新颖的交替训练策略,以减少多个领域之间的差异,同时概括为看不见的目标域。其次,在估计跨域差异测量时,我们提出了连续性约束。最后,为了使差异与迷你批量大小相结合,我们设计了一个特定领域的队列,以维护源域的全局视图,从而进一步提高了适应性性能。事实证明,我们的方法可以使用广泛的实验实现CDA问题的最新问题。该代码可在https://github.com/spiresearch/cda上找到。
translated by 谷歌翻译
最近,分布(OOD)的概括引起了人们对基于深度学习模型的鲁棒性和概括能力的关注,因此,已经制定了许多策略来解决与此问题相关的不同方面。但是,大多数现有的OOD概括算法都是复杂的,并且专门为某些数据集设计。为了减轻此问题,Nicochallenge-2022提供了Nico ++,这是一个具有不同上下文信息的大型数据集。在本文中,基于对NICO ++数据集的不同方案的系统分析,我们通过偶联的技巧提出了一个简单但有效的学习框架,包括多目标框架设计,数据增强,培训,培训和推理策略。我们的算法是记忆效率且易于安装的,没有复杂的模块,并且不需要大型预训练模型。它在公共测试集中获得了88.16%的前1位精度,在私人测试集中获得了75.65%的表现,并在域Nicochallenge-2022的域概括任务中排名第1。
translated by 谷歌翻译
肌肉骨骼和神经系统疾病是老年人行走问题的最常见原因,它们通常导致生活质量降低。分析步行运动数据手动需要训练有素的专业人员,并且评估可能并不总是客观的。为了促进早期诊断,最近基于深度学习的方法显示了自动分析的有希望的结果,这些方法可以发现传统的机器学习方法中未发现的模式。我们观察到,现有工作主要应用于单个联合特征,例如时间序列的联合职位。由于发现了诸如通常较小规模的医疗数据集的脚之间的距离(即步幅宽度)之类的挑战,因此这些方法通常是优选的。结果,我们提出了一种解决方案,该解决方案明确地将单个关节特征和关节间特征作为输入,从而使系统免于从小数据中发现更复杂的功能。由于两种特征的独特性质,我们引入了一个两流框架,其中一个流从关节位置的时间序列中学习,另一个从相对关节位移的时间序列中学习。我们进一步开发了一个中层融合模块,以将发现的两个流中发现的模式结合起来进行诊断,从而导致数据互补表示,以获得更好的预测性能。我们使用3D骨架运动的基准数据集涉及45例肌肉骨骼和神经系统疾病的患者,并实现95.56%的预测准确性,效果优于最先进的方法,从而验证了我们的系统。
translated by 谷歌翻译