We study the convergence of several natural policy gradient (NPG) methods in infinite-horizon discounted Markov decision processes with regular policy parametrizations. For a variety of NPGs and reward functions we show that the trajectories in state-action space are solutions of gradient flows with respect to Hessian geometries, based on which we obtain global convergence guarantees and convergence rates. In particular, we show linear convergence for unregularized and regularized NPG flows with the metrics proposed by Kakade and Morimura and co-authors by observing that these arise from the Hessian geometries of conditional entropy and entropy respectively. Further, we obtain sublinear convergence rates for Hessian geometries arising from other convex functions like log-barriers. Finally, we interpret the discrete-time NPG methods with regularized rewards as inexact Newton methods if the NPG is defined with respect to the Hessian geometry of the regularizer. This yields local quadratic convergence rates of these methods for step size equal to the penalization strength.
translated by 谷歌翻译
正如最近的作品中观察到的那样,通信图神经网络(GNN)中信号传播的质量强烈影响其表现力。特别是,对于依靠远程相互作用的预测任务,节点特征的递归聚合可能导致不希望的现象称为“过句”。我们提出了一个基于信息收缩的分析过度句子的框架。我们的分析以可靠计算的模型为指导,该模型由于冯·诺伊曼(Von Neumann),该模型在嘈杂的计算图中提供了新的洞察力作为信号淬灭的新见解。在此基础上,我们提出了一个旨在减轻过度量化的算法的图形。我们的算法采用了由扩展器图构造动机的随机局部边缘翻转原始的。我们将算法的光谱膨胀特性与现有基于曲率的非本地重新布线策略的光谱膨胀属性进行了比较。合成实验表明,尽管我们的算法通常具有较慢的膨胀速率,但总体计算更便宜,可以准确地保留节点度,并且永远不会断开图表的连接。
translated by 谷歌翻译
持续的同源性(PH)是拓扑数据分析中最流行的方法之一。尽管PH已用于许多不同类型的应用程序中,但其成功背后的原因仍然难以捉摸。特别是,尚不知道哪种类别的问题最有效,或者在多大程度上可以检测几何或拓扑特征。这项工作的目的是确定pH在数据分析中比其他方法更好甚至更好的问题。我们考虑三个基本形状分析任务:从形状采样的2D和3D点云中检测孔数,曲率和凸度。实验表明,pH在这些任务中取得了成功,超过了几个基线,包括PointNet,这是一个精确地受到点云的属性启发的体系结构。此外,我们观察到,pH对于有限的计算资源和有限的培训数据以及分布外测试数据,包括各种数据转换和噪声,仍然有效。
translated by 谷歌翻译
我们在假设目标函数的先前和EIGENExpansion系数的假定下,我们将高斯进程回归(GPR)的幂律渐近学习曲线的幂律渐近学呈现出高斯过程回归(GPR)。在类似的假设下,我们利用GPR和内核RIDGE回归(KRR)之间的等价性来显示KRR的泛化误差。无限宽的神经网络可以与GPR相对于神经网络GP内核和神经切线内核有关,其中已知在几个情况下具有幂律谱。因此,我们的方法可以应用于研究无限宽神经网络的泛化误差。我们提出了展示理论的玩具实验。
translated by 谷歌翻译
我们研究由线性卷积神经网络(LCN)代表的功能家族。这些函数形成了从输入空间到输出空间的线性地图集的半代数子集。相比之下,由完全连接的线性网络表示的函数家族形成代数集。我们观察到,LCN代表的功能可以通过接受某些因素化的多项式来识别,我们使用此视角来描述网络体系结构对所得功能空间几何形状的影响。我们进一步研究了在LCN上的目标函数的优化,分析了功能空间和参数空间中的临界点,并描述了梯度下降的动态不变性。总体而言,我们的理论预测,LCN的优化参数通常对应于跨层的重复过滤器,或可以分解为重复过滤器的过滤器。我们还进行了数值和符号实验,以说明我们的结果,并对小体系结构的景​​观进行深入分析。
translated by 谷歌翻译
使用神经网络学习依赖于可代表功能的复杂性,但更重要的是,典型参数的特定分配与不同复杂度的功能。将激活区域的数量作为复杂性度量,最近的作品表明,深度释放网络的实际复杂性往往远远远非理论最大值。在这项工作中,我们表明这种现象也发生在具有颤扬(多参数)激活功能的网络中,并且在考虑分类任务中的决策边界时。我们还表明参数空间具有多维全维区域,具有广泛不同的复杂性,并在预期的复杂性上获得非竞争下限。最后,我们调查了不同的参数初始化程序,并表明他们可以提高培训的收敛速度。
translated by 谷歌翻译
图形神经网络(GNNS)的表现力量受到限制,具有远程交互的斗争,缺乏模拟高阶结构的原则性方法。这些问题可以归因于计算图表和输入图结构之间的强耦合。最近提出的消息通过单独的网络通过执行图形的Clique复合物的消息来自然地解耦这些元素。然而,这些模型可能受到单纯复合物(SCS)的刚性组合结构的严重限制。在这项工作中,我们将最近的基于常规细胞复合物的理论结果扩展到常规细胞复合物,灵活地满满SCS和图表的拓扑物体。我们表明,该概括提供了一组强大的图表“提升”转换,每个图形是导致唯一的分层消息传递过程。我们集体呼叫CW Networks(CWNS)的结果方法比WL测试更强大,而不是比3 WL测试更强大。特别是,当应用于分子图问题时,我们证明了一种基于环的一个这样的方案的有效性。所提出的架构从可提供的较大的表达效益于常用的GNN,高阶信号的原则建模以及压缩节点之间的距离。我们展示了我们的模型在各种分子数据集上实现了最先进的结果。
translated by 谷歌翻译
我们介绍了可以由具有Maxout单位的人造馈电神经网络表示的功能线性区域的数量。排名kaxout单元是一个函数,计算$ k $线性函数的最大值。对于具有单层Maxout单元的网络,线性区域对应于Minkowski多型的上顶点。我们根据热带超曲面的交点或部分Minkowski总和的上面数,以及任何输入维度的区域数,任何单位数量,任何等级,任何等级,任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,以及任何等级,在有和没有偏见的情况下。基于这些结果,我们还为具有多层的网络获得了渐近的上限。
translated by 谷歌翻译
We study the complexity of functions computable by deep feedforward neural networks with piecewise linear activations in terms of the symmetries and the number of linear regions that they have. Deep networks are able to sequentially map portions of each layer's input-space to the same output. In this way, deep models compute functions that react equally to complicated patterns of different inputs. The compositional structure of these functions enables them to re-use pieces of computation exponentially often in terms of the network's depth. This paper investigates the complexity of such compositional maps and contributes new theoretical results regarding the advantage of depth for neural networks with piecewise linear activation functions. In particular, our analysis is not specific to a single family of models, and as an example, we employ it for rectifier and maxout networks. We improve complexity bounds from pre-existing work and investigate the behavior of units in higher layers.
translated by 谷歌翻译
Algorithms that involve both forecasting and optimization are at the core of solutions to many difficult real-world problems, such as in supply chains (inventory optimization), traffic, and in the transition towards carbon-free energy generation in battery/load/production scheduling in sustainable energy systems. Typically, in these scenarios we want to solve an optimization problem that depends on unknown future values, which therefore need to be forecast. As both forecasting and optimization are difficult problems in their own right, relatively few research has been done in this area. This paper presents the findings of the ``IEEE-CIS Technical Challenge on Predict+Optimize for Renewable Energy Scheduling," held in 2021. We present a comparison and evaluation of the seven highest-ranked solutions in the competition, to provide researchers with a benchmark problem and to establish the state of the art for this benchmark, with the aim to foster and facilitate research in this area. The competition used data from the Monash Microgrid, as well as weather data and energy market data. It then focused on two main challenges: forecasting renewable energy production and demand, and obtaining an optimal schedule for the activities (lectures) and on-site batteries that lead to the lowest cost of energy. The most accurate forecasts were obtained by gradient-boosted tree and random forest models, and optimization was mostly performed using mixed integer linear and quadratic programming. The winning method predicted different scenarios and optimized over all scenarios jointly using a sample average approximation method.
translated by 谷歌翻译