先前的作品已经为神经集功能建立了固体基础,以及有效的体系结构,这些架构保留了在集合上操作的必要属性,例如对集合元素的排列不变。随后,已经确定了在保持输出上保持一致性保证的同时,依次处理任何随机设置分区方案的任何置换的能力,但已建立了网络体系结构的选项有限。我们进一步研究了神经集编码功能中的MBC特性,建立了一种将任意非MBC模型转换为满足MBC的方法。在此过程中,我们为普遍MBC(UMBC)类的集合功能提供了一个框架。此外,我们探讨了通过我们的框架实现的有趣的辍学策略,并研究了其对测试时间分配变化下的概率校准的影响。我们通过单位测试支持的证据来验证UMBC,还提供了有关玩具数据,清洁和损坏的云云分类的定性/定量实验,并在Imagenet上摊销了聚类。结果表明了UMBC的实用性,我们进一步发现我们的辍学策略改善了不确定性校准。
translated by 谷歌翻译
Many machine learning tasks such as multiple instance learning, 3D shape recognition and fewshot image classification are defined on sets of instances. Since solutions to such problems do not depend on the order of elements of the set, models used to address them should be permutation invariant. We present an attention-based neural network module, the Set Transformer, specifically designed to model interactions among elements in the input set. The model consists of an encoder and a decoder, both of which rely on attention mechanisms. In an effort to reduce computational complexity, we introduce an attention scheme inspired by inducing point methods from sparse Gaussian process literature. It reduces computation time of self-attention from quadratic to linear in the number of elements in the set. We show that our model is theoretically attractive and we evaluate it on a range of tasks, demonstrating increased performance compared to recent methods for set-structured data.
translated by 谷歌翻译
Learning object-centric representations of complex scenes is a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep learning approaches learn distributed representations that do not capture the compositional properties of natural scenes. In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots. These slots are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention. We empirically demonstrate that Slot Attention can extract object-centric representations that enable generalization to unseen compositions when trained on unsupervised object discovery and supervised property prediction tasks.
translated by 谷歌翻译
大多数设置深度学习的预测模型,使用Set-Scifariant操作,但它们实际上在MultiSet上运行。我们表明设置的函数不能代表多种功能上的某些功能,因此我们介绍了更适当的多种式概念概念。我们确定现有的深度设置预测网络(DSPN)可以是多机构的,而不会被设定的标准规模阻碍,并通过近似隐式差分改进它,允许更好地优化,同时更快和节省存储器。在一系列玩具实验中,我们表明,多机构的角度是有益的,在大多数情况下,我们对DSPN的变化达到了更好的结果。关于CLEVR对象性质预测,由于通过隐含分化所取得的益处,我们在最先进的评估指标中从8%到77%的最先进的槽注意力从8%提高到77%。
translated by 谷歌翻译
We study the problem of designing models for machine learning tasks defined on sets. In contrast to traditional approach of operating on fixed dimensional vectors, we consider objective functions defined on sets that are invariant to permutations. Such problems are widespread, ranging from estimation of population statistics [1], to anomaly detection in piezometer data of embankment dams [2], to cosmology [3,4]. Our main theorem characterizes the permutation invariant functions and provides a family of functions to which any permutation invariant objective function must belong. This family of functions has a special structure which enables us to design a deep network architecture that can operate on sets and which can be deployed on a variety of scenarios including both unsupervised and supervised learning tasks. We also derive the necessary and sufficient conditions for permutation equivariance in deep models. We demonstrate the applicability of our method on population statistic estimation, point cloud classification, set expansion, and outlier detection.
translated by 谷歌翻译
我们为在多个置换不变的集合上学习功能提出了一个一般的深度体系结构。我们还展示了如何通过维度等值的任何维度元素概括到任何维度元素的集合。我们证明了我们的体系结构是这些功能的通用近似值,并显示了有关各种任务的现有方法的卓越结果,包括计数任务,对齐任务,可区分性任务和统计距离测量。最后的任务在机器学习中非常重要。尽管我们的方法非常笼统,但我们证明它可以产生KL差异和相互信息的近似估计值,这些信息比以前专门设计以近似这些统计距离的技术更准确。
translated by 谷歌翻译
Accurate uncertainty quantification is a major challenge in deep learning, as neural networks can make overconfident errors and assign high confidence predictions to out-of-distribution (OOD) inputs. The most popular approaches to estimate predictive uncertainty in deep learning are methods that combine predictions from multiple neural networks, such as Bayesian neural networks (BNNs) and deep ensembles. However their practicality in real-time, industrial-scale applications are limited due to the high memory and computational cost. Furthermore, ensembles and BNNs do not necessarily fix all the issues with the underlying member networks. In this work, we study principled approaches to improve uncertainty property of a single network, based on a single, deterministic representation. By formalizing the uncertainty quantification as a minimax learning problem, we first identify distance awareness, i.e., the model's ability to quantify the distance of a testing example from the training data, as a necessary condition for a DNN to achieve high-quality (i.e., minimax optimal) uncertainty estimation. We then propose Spectral-normalized Neural Gaussian Process (SNGP), a simple method that improves the distance-awareness ability of modern DNNs with two simple changes: (1) applying spectral normalization to hidden weights to enforce bi-Lipschitz smoothness in representations and (2) replacing the last output layer with a Gaussian process layer. On a suite of vision and language understanding benchmarks, SNGP outperforms other single-model approaches in prediction, calibration and out-of-domain detection. Furthermore, SNGP provides complementary benefits to popular techniques such as deep ensembles and data augmentation, making it a simple and scalable building block for probabilistic deep learning. Code is open-sourced at https://github.com/google/uncertainty-baselines
translated by 谷歌翻译
置换不变的神经网络是从集合进行预测的有前途的工具。但是,我们表明,现有的置换式体系结构,深度集和固定的变压器可能会在深度时消失或爆炸。此外,层规范(SET变压器中选择的归一化)可能会通过删除对预测有用的信息来损害性能。为了解决这些问题,我们介绍了白皮剩余连接的干净路径原理,并开发了设置规范,这是针对集合量身定制的标准化。有了这些,我们构建了Deep Sets ++和SET Transformer ++,该模型比其在各种任务套件上的原始配对品具有可比性或更好的性能。我们还引入了Flow-RBC,这是一种新的单细胞数据集和置换不变预测的现实应用。我们在此处开放数据和代码:https://github.com/rajesh-lab/deep_permunt_invariant。
translated by 谷歌翻译
现有的置换不变方法可以根据聚合范围(即全球聚合和局部局部)分为两类。尽管全局聚合方法,e。 g。,PointNet和Deep Sets,参与更简单的结构,它们的性能比PointNet ++和Point Transformer等局部聚合较差。如果存在具有简单结构,竞争性能甚至更少参数的全球聚合方法,那么它仍然是一个空旷的问题。在本文中,我们提出了一个基于双MLP点产品的新型全局聚合置换不变的网络,称为DUMLP-PIN,该网络能够用于提取集合输入的功能,包括无序或非结构的像素,属性,atter和Point和Point和Point云数据集。我们严格地证明,DUMLP-PIN实现的任何置换不变函数都可以通过点产生方式分解为两个或多个置换量的函数,因为给定输入集的基数大于阈值。我们还表明,在某些条件下,可以将DUMLP针视为具有强大限制的深度集。 DUMLP-PIN的性能在具有不同数据集的几个不同任务上进行了评估。实验结果表明,我们的DUMLP-PIN在像素集和属性集的两个分类问题上取得了最佳结果。在点云分类和零件分割上,DUMLP-PIN的准确性非常接近SO-FAR最佳表现最佳的本地聚合方法,仅差异1-2%,而所需参数的数量显着降低了分类分别超过85%和69%的分割。该代码可在https://github.com/jaronthu/dumlp-pin上公开获得。
translated by 谷歌翻译
Continual Learning (CL) is a field dedicated to devise algorithms able to achieve lifelong learning. Overcoming the knowledge disruption of previously acquired concepts, a drawback affecting deep learning models and that goes by the name of catastrophic forgetting, is a hard challenge. Currently, deep learning methods can attain impressive results when the data modeled does not undergo a considerable distributional shift in subsequent learning sessions, but whenever we expose such systems to this incremental setting, performance drop very quickly. Overcoming this limitation is fundamental as it would allow us to build truly intelligent systems showing stability and plasticity. Secondly, it would allow us to overcome the onerous limitation of retraining these architectures from scratch with the new updated data. In this thesis, we tackle the problem from multiple directions. In a first study, we show that in rehearsal-based techniques (systems that use memory buffer), the quantity of data stored in the rehearsal buffer is a more important factor over the quality of the data. Secondly, we propose one of the early works of incremental learning on ViTs architectures, comparing functional, weight and attention regularization approaches and propose effective novel a novel asymmetric loss. At the end we conclude with a study on pretraining and how it affects the performance in Continual Learning, raising some questions about the effective progression of the field. We then conclude with some future directions and closing remarks.
translated by 谷歌翻译
内核选择在确定高斯过程(GP)模型中的性能方面发挥着核心作用,因为所选择的内核在之前的GP下确定了电感偏差和在GP下的功能的先前支持。这项工作解决了为高维GP回归模型构建自定义内核功能的挑战。从最近的深度学习进步中汲取灵感,我们介绍了一个名为Kitt的新方法:通过变压器识别内核识别。 KITT利用基于变压器的架构,以在0.1秒内生成内核建议,这比传统的内核搜索算法快几个数量级。我们使用从已知内核的词汇表中从前线生成的合成数据训练我们的模型。通过利用自我关注机制的性质,KITT能够处理具有任意尺寸的输入的数据集。我们证明,KITT选择的内核会在各种回归基准集合中产生强烈的表现。
translated by 谷歌翻译
学习神经集功能在许多应用中越来越重要,例如产品推荐和AI辅助药物发现中的复合选择。在功能值Oracle下,大多数现有的作品研究方法学方法学方法学都需要昂贵的监督信号。这使得仅在最佳子集(OS)Oracle下仅进行弱监督的应用程序使其不切实际,而研究的研究令人惊讶地忽略了。在这项工作中,我们提出了一个原则上但实用的最大似然学习框架,称为等效性,该框架同时满足OS ORACLE下的以下学习设置功能:i)置入了模型的设定质量函数的置换率; ii)许可不同地面套件; iii)最低先验;和iv)可伸缩性。我们框架的主要组成部分涉及:对设定质量函数的基于能量的处理,深空式体系结构来处理置换不变性,平均场变异推理及其摊销变体。由于这些高级体系结构的优雅组合,对三个现实世界应用的实证研究(包括亚马逊产品推荐,设置异常检测和虚拟筛选的复合选择)表明,EquivSet的表现优于基本线的大幅度。
translated by 谷歌翻译
神经过程(NP)是一种流行的元学习方法。与高斯工艺(GPS)类似,NPS将分布定义在功能上,并可以估计其预测中的不确定性。但是,与GPS不同,NP及其变体遭受不足的折磨,并且通常具有棘手的可能性,这限制了其在顺序决策中的应用。我们提出了变形金刚神经过程(TNP),这是NP家族的新成员,将不确定性感知的元学习作为序列建模问题。我们通过基于自回旋的可能性目标学习TNP,并通过新颖的基于变压器的建筑实例化。该模型架构尊重问题结构固有的归纳偏差,例如对观察到的数据点的不变性以及与未观察到的点的等效性。我们进一步研究了TNP框架内的旋钮,以额外的计算来折衷解码分布的表达。从经验上讲,我们表明TNP在各种基准问题上实现最新性能,在元回归,图像完成,上下文多武器匪徒和贝叶斯优化方面表现优于所有先前的NP变体。
translated by 谷歌翻译
学习将输入集映射到其元素元素序列上的任务对于神经网络而言是一项挑战。设置到序列问题发生在自然语言处理,计算机视觉和结构预测中,其中大集合元素之间的相互作用定义了最佳输出。模型必须表现出关系推理,处理不同的基础性并管理组合复杂性。以前的基于注意力的方法需要$ n $层的设定转换,以明确表示$ n $ th订单关系。我们的目的是增强他们通过附加相互依赖组件有效地对高阶相互作用进行有效建模的能力。我们提出了一种新型的神经集编码方法,称为“集合相互依赖变压器”,能够将集合的置换不变表示与其在任何基数集合中的元素联系起来。我们将其与置换学习模块结合到一个完整的三部分设定模型中,并在许多任务上演示其最先进的性能。这些范围从组合优化问题,到在合成和已建立的NLP数据集上的置换学习挑战到句子排序的挑战,到产品目录结构预测的新颖领域。此外,研究了网络概括到看不见的序列长度的能力,并提供了对现有方法学习高阶相互作用能力的比较经验分析。
translated by 谷歌翻译
以对象为中心的表示是通过提供柔性抽象可以在可以建立的灵活性抽象来实现更系统的推广的有希望的途径。最近的简单2D和3D数据集的工作表明,具有对象的归纳偏差的模型可以学习段,并代表单独的数据的统计结构中的有意义对象,而无需任何监督。然而,尽管使用越来越复杂的感应偏差(例如,用于场景的尺寸或3D几何形状),但这种完全无监督的方法仍然无法扩展到不同的现实数据。在本文中,我们采取了弱监督的方法,并专注于如何使用光流的形式的视频数据的时间动态,2)调节在简单的对象位置上的模型可以用于启用分段和跟踪对象在明显更现实的合成数据中。我们介绍了一个顺序扩展,以便引入我们训练的推出,我们训练用于预测现实看的合成场景的光流,并显示调节该模型的初始状态在一小组提示,例如第一帧中的物体的质量中心,是足以显着改善实例分割。这些福利超出了新型对象,新颖背景和更长的视频序列的培训分配。我们还发现,在推论期间可以使用这种初始状态调节作为对特定物体或物体部分的型号查询模型,这可能会为一系列弱监管方法铺平,并允许更有效的互动训练有素的型号。
translated by 谷歌翻译
通常通过将许多输入张量汇总为单个表示形式来处理神经网络中神经网络中的处理集或其他无序的,潜在的变化大小的输入。尽管从简单的汇总到多头关注已经存在许多聚合方法,但从理论和经验的角度来看,它们的代表力都受到限制。在搜索主要功能更强大的聚合策略时,我们提出了一种基于优化的方法,称为平衡聚​​集。我们表明,许多现有的聚合方法可以作为平衡聚集的特殊情况恢复,并且在某些重要情况下,它效率更高。在许多现有的架构和应用中,平衡聚集可以用作置换式替换。我们在三个不同的任务上验证其效率:中值估计,班级计数和分子性质预测。在所有实验中,平衡聚集的性能都比我们测试的其他聚合技术更高。
translated by 谷歌翻译
收购用于监督学习的标签可能很昂贵。为了提高神经网络回归的样本效率,我们研究了活跃的学习方法,这些方法可以适应地选择未标记的数据进行标记。我们提出了一个框架,用于从(与网络相关的)基础内核,内核转换和选择方法中构造此类方法。我们的框架涵盖了许多基于神经网络的高斯过程近似以及非乘式方法的现有贝叶斯方法。此外,我们建议用草图的有限宽度神经切线核代替常用的最后层特征,并将它们与一种新型的聚类方法结合在一起。为了评估不同的方法,我们引入了一个由15个大型表格回归数据集组成的开源基准。我们所提出的方法的表现优于我们的基准测试上的最新方法,缩放到大数据集,并在不调整网络体系结构或培训代码的情况下开箱即用。我们提供开源代码,包括所有内核,内核转换和选择方法的有效实现,并可用于复制我们的结果。
translated by 谷歌翻译
In this work, we present Point Transformer, a deep neural network that operates directly on unordered and unstructured point sets. We design Point Transformer to extract local and global features and relate both representations by introducing the local-global attention mechanism, which aims to capture spatial point relations and shape information. For that purpose, we propose SortNet, as part of the Point Transformer, which induces input permutation invariance by selecting points based on a learned score. The output of Point Transformer is a sorted and permutation invariant feature list that can directly be incorporated into common computer vision applications. We evaluate our approach on standard classification and part segmentation benchmarks to demonstrate competitive results compared to the prior work. Code is publicly available at: https://github.com/engelnico/point-transformer INDEX TERMS 3D point processing, Artificial neural networks, Computer vision, Feedforward neural networks, Transformer
translated by 谷歌翻译
Deep neural nets with a large number of parameters are very powerful machine learning systems. However, overfitting is a serious problem in such networks. Large networks are also slow to use, making it difficult to deal with overfitting by combining the predictions of many different large neural nets at test time. Dropout is a technique for addressing this problem. The key idea is to randomly drop units (along with their connections) from the neural network during training. This prevents units from co-adapting too much. During training, dropout samples from an exponential number of different "thinned" networks. At test time, it is easy to approximate the effect of averaging the predictions of all these thinned networks by simply using a single unthinned network that has smaller weights. This significantly reduces overfitting and gives major improvements over other regularization methods. We show that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
translated by 谷歌翻译
在本文中,我们涉及在2D点云数据上的旋转设备。我们描述了一种特定的功能,能够近似任何连续旋转等级和置换不变函数。基于这一结果,我们提出了一种新的神经网络架构,用于处理2D点云,我们证明其普遍性地用于近似呈现这些对称的功能。我们还展示了如何扩展架构以接受一组2D-2D对应关系作为Indata,同时保持类似的标准性属性。关于立体视觉中必需基质的估计的实验。
translated by 谷歌翻译