We propose a new class of implicit networks, the multiscale deep equilibrium model (MDEQ), suited to large-scale and highly hierarchical pattern recognition domains. An MDEQ directly solves for and backpropagates through the equilibrium points of multiple feature resolutions simultaneously, using implicit differentiation to avoid storing intermediate states (and thus requiring only $O(1)$ memory consumption). These simultaneously-learned multi-resolution features allow us to train a single model on a diverse set of tasks and loss functions, such as using a single MDEQ to perform both image classification and semantic segmentation. We illustrate the effectiveness of this approach on two large-scale vision tasks: ImageNet classification and semantic segmentation on high-resolution images from the Cityscapes dataset. In both settings, MDEQs are able to match or exceed the performance of recent competitive computer vision models: the first time such performance and scale have been achieved by an implicit deep learning approach. The code and pre-trained models are at https://github.com/locuslab/mdeq .
Janelia Farm Research Campus（珍妮莉亚农场研究园区）；Howard Hughes Medical Institute（霍华德·休斯医学研究所）
In this work, we propose a learning framework for identifying synapses using a deep and wide multi-scale recursive (DAWMR) network, previously considered in image segmentation applications. We apply this approach on electron microscopy data from invertebrate fly brain tissue. By learning features directly from the data, we are able to achieve considerable improvements over existing techniques that rely on a small set of hand-designed features. We show that this system can reduce the amount of manual annotation required, in both acquisition of training data as well as verification of inferred d…
Visual saliency is a fundamental problem in both cognitive and computational sciences, including computer vision. In this CVPR 2015 paper, we discover that a high-quality visual saliency model can be trained with multiscale features extracted using a popular deep learning architecture, convolutional neural networks (CNNs), which have had many successes in visual recognition tasks. For learning such saliency models, we introduce a neural network architecture, which has fully connected layers on top of CNNs responsible for extracting features at three different scales. We then propose a refineme…
In this paper, we propose a simple but effective message passing method to improve the boundary quality for the semantic segmentation result. Inspired by the generated sharp edges of superpixel blocks, we employ superpixel to guide the information passing within feature map. Simultaneously, the sharp boundaries of the blocks also restrict the message passing scope. Specifically, we average features that the superpixel block covers within feature map, and add the result back to each feature vector. Further, to obtain sharper edges and farther spatial dependence, we develop a multiscale superpix…
Xinjiang University（新疆大学）；Software College, Xinjiang University（新疆大学软件学院）；Network Center, Xinjiang University（新疆大学网络中心）
To better retain the deep features of an image and solve the sparsity problem of the end-to-end segmentation model, we propose a new deep convolutional network model for medical image pixel segmentation, called MC-Net. The core of this network model consists of four parts, namely, an encoder network, a multiple max-pooling integration module, a cross multiscale deconvolution decoder network and a pixel-level classification layer. In the network structure of the encoder, we use multiscale convolution instead of the traditional single-channel convolution. The multiple max-pooling integration mod…
Although much significant progress has been made in the research field of object detection with deep learning, there still exists a challenging task for the objects with small size, which is notably pronounced in UAV-captured images. Addressing these issues, it is a critical need to explore the feature extraction methods that can extract more sufficient feature information of small objects. In this paper, we propose a novel method called Dense Multiscale Feature Fusion Pyramid Networks(DMFFPN), which is aimed at obtaining rich features as much as possible, improving the information propagation…
U3M: Unbiased Multiscale Modal Fusion Model for Multimodal Semantic Segmentation
Institute of Artificial Intelligence (TeleAI), China Telecom（中国电信人工智能研究院（TeleAI））；School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University（西北工业大学人工智能、光学与电子学院（iOPEN））
Localization of chest pathologies in chest X-ray images is a challenging task because of their varying sizes and appearances. We propose a novel weakly supervised method to localize chest pathologies using class aware deep multiscale feature learning. Our method leverages intermediate feature maps from CNN layers at different stages of a deep network during the training of a classification model using image level annotations of pathologies. During the training phase, a set of \emph{layer relevance weights} are learned for each pathology class and the CNN is optimized to perform pathology class…
MSDS: Deep Structural Similarity with Multiscale Representation
Fujian Key Lab for Intelligent Processing and Wireless Communication of Media Information, College of Physics and Information Engineering, Fuzhou University（福建省智能信息处理与无线通信重点实验室，物理与信息工程学院，福州大学）；School of Computing and Information Science, Fuzhou Institute of Technology（计算机与信息科学学院，福州理工学院）
Fine-scale simulation of complex systems governed by multiscale partial differential equations (PDEs) is computationally expensive and various multiscale methods have been developed for addressing such problems. In addition, it is challenging to develop accurate surrogate and uncertainty quantification models for high-dimensional problems governed by stochastic multiscale PDEs using limited training data. In this work to address these challenges, we introduce a novel hybrid deep-learning and multiscale approach for stochastic multiscale PDEs with limited training data. For demonstration purpos…
Echo state networks are computationally lightweight reservoir models inspired by the random projections observed in cortical circuitry. As interest in reservoir computing has grown, networks have become deeper and more intricate. While these networks are increasingly applied to nontrivial forecasting tasks, there is a need for comprehensive performance analysis of deep reservoirs. In this work, we study the influence of partitioning neurons given a budget and the effect of parallel reservoir pathways across different datasets exhibiting multi-scale and nonlinear dynamics.
Deep Neural Networks (DNNs) provide state-of-the-art solutions in several difficult machine perceptual tasks. However, their performance relies on the availability of a large set of labeled training data, which limits the breadth of their applicability. Hence, there is a need for new {\em semi-supervised learning} methods for DNNs that can leverage both (a small amount of) labeled and unlabeled training data. In this paper, we develop a general loss function enabling DNNs of any topology to be trained in a semi-supervised manner without extra hyper-parameters. As opposed to current semi-superv…
University of Amsterdam（阿姆斯特丹大学）；Ecole Normale Supérieure（巴黎高等师范学院）
Deep neural network algorithms are difficult to analyze because they lack structure allowing to understand the properties of underlying transforms and invariants. Multiscale hierarchical convolutional networks are structured deep convolutional networks where layers are indexed by progressively higher dimensional attributes, which are learned from training data. Each new layer is computed with multidimensional convolutions along spatial and attribute variables. We introduce an efficient implementation of such networks where the dimensionality is progressively reduced by averaging intermediate l…
In this paper we are interested in the problem of learning an over-complete basis and a methodology such that the reconstruction or inverse problem does not need optimization. We analyze the optimality of the presented approaches, their link to popular already known techniques s.a. Artificial Neural Networks,k-means or Oja's learning rule. Finally, we will see that one approach to reach the optimal dictionary is a factorial and hierarchical approach. The derived approach lead to a formulation of a Deep Oja Network. We present results on different tasks and present the resulting very efficient …
ShanghaiTech University（上海科技大学）；University of Notre Dame（圣母大学）；Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences（中国科学院上海微系统与信息技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Estimation of spatially-varying parameters for computationally expensive forward models governed by partial differential equations is addressed. A novel multiscale Bayesian inference approach is introduced based on deep probabilistic generative models. Such generative models provide a flexible representation by inferring on each scale a low-dimensional latent encoding while allowing hierarchical parameter generation from coarse- to fine-scales. Combining the multiscale generative model with Markov Chain Monte Carlo (MCMC), inference across scales is achieved enabling us to efficiently obtain p…
Discovering Multiscale Deep Formulas in Complex Systems via Neural-Guided Lambda Calculus
National Engineering Laboratory for Big Data Analytics（大数据分析国家工程实验室）；Xi’an Jiaotong University（西安交通大学）；School of Mathematics and Statistics（数学与统计学学院）；School of Computer Science and Technology, Faculty of Electronic and Information Engineering（计算机科学与技术学院，电子与信息工程学院）
Multiscale modeling is an effective approach for investigating multiphysics systems with largely disparate size features, where models with different resolutions or heterogeneous descriptions are coupled together for predicting the system's response. The solver with lower fidelity (coarse) is responsible for simulating domains with homogeneous features, whereas the expensive high-fidelity (fine) model describes microscopic features with refined discretization, often making the overall cost prohibitively high, especially for time-dependent problems. In this work, we explore the idea of multisca…
Janelia Farm Research Campus（珍妮莉亚法姆研究园区）；Howard Hughes Medical Institute（霍华德·休斯医学研究所）
Feedforward multilayer networks trained by supervised learning have recently demonstrated state of the art performance on image labeling problems such as boundary prediction and scene parsing. As even very low error rates can limit practical usage of such systems, methods that perform closer to human accuracy remain desirable. In this work, we propose a new type of network with the following properties that address what we hypothesize to be limiting aspects of existing methods: (1) a `wide' structure with thousands of features, (2) a large field of view, (3) recursive iterations that exploit s…
INSA Centre Val de Loire（卢瓦尔河谷国立应用科学学院）；PRISME（PRISME实验室）
This paper addresses the problem of liver cancer segmentation in Whole Slide Image (WSI). We propose a multi-scale image processing method based on automatic end-to-end deep neural network algorithm for segmentation of cancer area. A seven-levels gaussian pyramid representation of the histopathological image was built to provide the texture information in different scales. In this work, several neural architectures were compared using the original image level for the training procedure. The proposed method is based on U-Net applied to seven levels of various resolutions (pyramidal subsumpling)…
Despite the increasing importance of strain localization modeling (e.g., failure analysis) in computer-aided engineering, there is a lack of effective approaches to capturing relevant material behaviors consistently across multiple length scales. We aim to address this gap within the framework of deep material networks (DMN) -- a machine learning model with embedded mechanics in the building blocks. A new cell-division scheme is proposed to track the scale transition through the network, and its consistency is ensured by the physics of fitting parameters. Essentially, each microscale node in t…
Université Paris-Saclay（巴黎-萨克雷大学）；CNRS（法国国家科学研究中心）；CentraleSupelec（中央理工-高等电力学院）；Laboratoire des Signaux et Systèmes（信号与系统实验室）
We propose a practical deep generative approach for lossless point cloud geometry compression, called MSVoxelDNN, and show that it significantly reduces the rate compared to the MPEG G-PCC codec. Our previous work based on autoregressive models (VoxelDNN) has a fast training phase, however, inference is slow as the occupancy probabilities are predicted sequentially, voxel by voxel. In this work, we employ a multiscale architecture which models voxel occupancy in coarse-to-fine order. At each scale, MSVoxelDNN divides voxels into eight conditionally independent groups, thus requiring a single n…
Climate change impacts could cause progressive decrease of crop quality and yield, up to harvest failures. In particular, heat waves and other climate extremes can lead to localized food shortages and even threaten food security of communities worldwide. In this study, we apply a deep learning architecture for high resolution forecasting (300 m, 10 days) of the Leaf Area Index (LAI), whose dynamics has been widely used to model the growth phase of crops and impact of heat waves. LAI models can be computed at 0.1 degree spatial resolution with an auto regressive component adjusted with weather …
This paper extends the deep material network (DMN) proposed by Liu et al. (2019) to tackle general 3-dimensional (3D) problems with arbitrary material and geometric nonlinearities. It discovers a new way of describing multiscale heterogeneous materials by a multi-layer network structure and mechanistic building blocks. The data-driven framework of DMN is discussed in detail about the offline training and online extrapolation stages. Analytical solutions of the 3D building block with a two-layer structure in both small- and finite-strain formulations are derived based on interfacial equilibrium…
This paper proposes multiscale convolutional neural network (CNN)-based deep metric learning for bioacoustic classification, under low training data conditions. The proposed CNN is characterized by the utilization of four different filter sizes at each level to analyze input feature maps. This multiscale nature helps in describing different bioacoustic events effectively: smaller filters help in learning the finer details of bioacoustic events, whereas, larger filters help in analyzing a larger context leading to global details. A dynamic triplet loss is employed in the proposed CNN architectu…
Nonlinear differential equations rarely admit closed-form solutions, thus requiring numerical time-stepping algorithms to approximate solutions. Further, many systems characterized by multiscale physics exhibit dynamics over a vast range of timescales, making numerical integration computationally expensive due to numerical stiffness. In this work, we develop a hierarchy of deep neural network time-steppers to approximate the flow map of the dynamical system over a disparate range of time-scales. The resulting model is purely data-driven and leverages features of the multiscale dynamics, enabli…
EquiNO: A Physics-Informed Neural Operator for Multiscale Simulations
Institute for Software and Systems Engineering, Clausthal University of Technology（软件与系统工程研究所，克劳斯塔尔大学）；Institute of Applied Mechanics, Clausthal University of Technology（应用力学研究所，克劳斯塔尔大学）
Department of Mathematics, Texas A&M University（德克萨斯大学数学系）；Department of Mathematics, The Chinese University of Hong Kong（香港中文大学数学系）；Department of Mathematics & Institute for Scientific Computation (ISC), Texas A&M University（德克萨斯大学数学系与科学计算研究所）
Stanford University（斯坦福大学）；University of Dundee（邓迪大学）；University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）
Timely completion of design cycles for complex systems ranging from consumer electronics to hypersonic vehicles relies on rapid simulation-based prototyping. The latter typically involves high-dimensional spaces of possibly correlated control variables (CVs) and quantities of interest (QoIs) with non-Gaussian and possibly multimodal distributions. We develop a model-agnostic, moment-independent global sensitivity analysis (GSA) that relies on differential mutual information to rank the effects of CVs on QoIs. The data requirements of this information-theoretic approach to GSA are met by replac…
Structure and asymptotic preserving deep neural surrogates for uncertainty quantification in multiscale kinetic equations
University of Ferrara（费拉拉大学）；Italian Ministry of University and Research（意大利大学与研究部）；Royal Society（皇家学会）；European Union（欧盟）；ICSC – Centro Nazionale di Ricerca in High Performance Computing, Big Data and Quantum Computing（ICSC – 高性能计算、大数据和量子计算研究中心）；Italian National Institute of High Mathematics（意大利国家高等数学研究所）
The history-dependent behaviors of classical plasticity models are often driven by internal variables evolved according to phenomenological laws. The difficulty to interpret how these internal variables represent a history of deformation, the lack of direct measurement of these internal variables for calibration and validation, and the weak physical underpinning of those phenomenological laws have long been criticized as barriers to creating realistic models. In this work, geometric machine learning on graph data (e.g. finite element solutions) is used as a means to establish a connection betw…
Facebook AI Research（Facebook AI研究院）；UC Berkeley（加州大学伯克利分校）
We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, …
In recent years, several point cloud geometry compression methods that utilize advanced deep learning techniques have been proposed, but there are limited works on attribute compression, especially lossless compression. In this work, we build an end-to-end multiscale point cloud attribute coding method (MNeT) that progressively projects the attributes onto multiscale latent spaces. The multiscale architecture provides an accurate context for the attribute probability modeling and thus minimizes the coding bitrate with a single network prediction. Besides, our method allows scalable coding that…
