Neural networks have recently allowed solving many ill-posed inverse problems with unprecedented performance. Physics informed approaches already progressively replace carefully hand-crafted reconstruction algorithms in real applications. However, these networks suffer from a major defect: when trained on a given forward operator, they do not generalize well to a different one. The aim of this paper is twofold. First, we show through various applications that training the network with a family of forward operators allows solving the adaptivity problem without compromising the reconstruction quality significantly. Second, we illustrate that this training procedure allows tackling challenging blind inverse problems. Our experiments include partial Fourier sampling problems arising in magnetic resonance imaging (MRI), computerized tomography (CT) and image deblurring.
translated by 谷歌翻译
事件摄像机由于其有益的特性,例如高时间分辨率,高带宽,几乎没有运动模糊和低功耗,因此在机器人技术和计算机视觉中变得越来越流行。但是,这些相机在市场上仍然昂贵且稀缺,使它们无法获得大多数。使用事件模拟器最大程度地减少了对真实事件摄像机开发新算法的需求。但是,由于模拟的计算复杂性,无法实时生成现有仿真器的事件流,而是必须从现有视频序列或预渲染中预先计算,然后从虚拟3D场景中进行模拟。尽管这些离线生成的事件流可以用作学习任务的培训数据,但所有响应时间的应用程序都无法从这些模拟器中受益,因为它们仍然需要实际的事件摄像头。这项工作提出了仿真方法,将事件模拟的性能提高了两个数量级(使其实时能够),同时在质量评估中保持竞争力。
translated by 谷歌翻译
在稀疏的奖励设置学习最优策略是困难的,因为学习代理人也鲜有其行动的质量没有反馈。在这些情况下,一个好的策略是专注于探索,希望能导致回报信号,以改善的发现。一个能够处理这种设置的学习算法必须能够(1)探讨可能的代理行为和(2)利用任何可能发现的奖励。高效勘探算法已经被提出,需要在被称为是一个值得探讨的空间中定义一个行为空间,即联营公司代理其产生的行为。需要定义这个空间是这些算法的限制。在这项工作中,我们介绍了STAX,旨在学习上的即时行为空间,并探索它的同时有效地优化发现任何报酬的算法。它通过分离的探索,并通过交替的两步过程中从奖励的剥削行为空间的学习这样做。在第一步骤中,建立STAX多样化策略的所有组成成分,同时学习策略评估过程中产生的高维观测值的低维表示。在开发步骤中,发射器用于优化发现有价值的解决方案的性能。在三个不同的稀疏奖励的环境进行的实验显示,STAX执行同等于现有基准,同时要求有关任务的要少得多的先验信息,因为它建立自主的行为空间。
translated by 谷歌翻译
在最近的方法论文中,我们展示了如何使用当地集合卡尔曼滤波器来学习混沌动力学以及状态轨迹。在这里,我们更系统地调查使用具有协方差定位或本地域的本地集合卡尔曼滤波器的可能性,以便检索状态和密钥全局和本地参数的混合。全局参数旨在代表代理动态核心,例如通过神经网络,这些核心让人想起数据驱动的动态机器学习,而本地参数通常代表模型的强制。针对联合状态和参数估计,提出了一种用于协方差和局域定位的一系列算法。特别是,我们展示了如何使用诸如本地集合变换卡尔曼滤波器(LetkF),这是一个固有的本地方法的本地域集合Kalman滤波器(ENKF)严格更新全局参数。使用几种本地ENKF味道在40变量LORENZ模型上取得了成功测试方法。最终提供基于多层Lorenz模型的二维图示。它使用辐射状的非本地观测。它具有本地域名和协方差本地化,以便学习混沌动态和本地强制。本文始终涉及全局和本地模型参数的在线估计的关键问题。
translated by 谷歌翻译
Deep learning frameworks have often focused on either usability or speed, but not both. PyTorch is a machine learning library that shows that these two goals are in fact compatible: it provides an imperative and Pythonic programming style that supports code as a model, makes debugging easy and is consistent with other popular scientific computing libraries, while remaining efficient and supporting hardware accelerators such as GPUs. In this paper, we detail the principles that drove the implementation of PyTorch and how they are reflected in its architecture. We emphasize that every aspect of PyTorch is a regular Python program under the full control of its user. We also explain how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance. We demonstrate the efficiency of individual subsystems, as well as the overall speed of PyTorch on several common benchmarks.
translated by 谷歌翻译