October 20, 2019

3173 words 15 mins read

Paper Group AWR 266

Understanding Deep Architectures by Visual Summaries. Support Neighbor Loss for Person Re-Identification. fastMRI: An Open Dataset and Benchmarks for Accelerated MRI. How Many Random Seeds? Statistical Power Analysis in Deep Reinforcement Learning Experiments. Preconditioner on Matrix Lie Group for SGD. FingerVision Tactile Sensor Design and Slip D …

Understanding Deep Architectures by Visual Summaries


Title	Understanding Deep Architectures by Visual Summaries
Authors	Marco Carletti, Marco Godi, Maedeh Aghaei, Francesco Giuliari, Marco Cristani
Abstract	In deep learning, visualization techniques extract the salient patterns exploited by deep networks for image classification, focusing on single images; no effort has been spent in investigating whether these patterns are systematically related to precise semantic entities over multiple images belonging to a same class, thus failing to capture the very understanding of the image class the network has realized. This paper goes in this direction, presenting a visualization framework which produces a group of clusters or summaries, each one formed by crisp salient image regions focusing on a particular part that the network has exploited with high regularity to decide for a given class. The approach is based on a sparse optimization step providing sharp image saliency masks that are clustered together by means of a semantic flow similarity measure. The summaries communicate clearly what a network has exploited of a particular image class, and this is proved through automatic image tagging and with a user study. Beyond the deep network understanding, summaries are also useful for many quantitative reasons: their number is correlated with ability of a network to classify (more summaries, better performances), and they can be used to improve the classification accuracy of a network through summary-driven specializations.
Tasks	Image Classification
Published	2018-01-27
URL	https://arxiv.org/abs/1801.09103v3
PDF	https://arxiv.org/pdf/1801.09103v3.pdf
PWC	https://paperswithcode.com/paper/understanding-deep-architectures-by-visual
Repo	https://github.com/godimarcovr/interpretable_visual_summaries
Framework	none

Support Neighbor Loss for Person Re-Identification


Title	Support Neighbor Loss for Person Re-Identification
Authors	Kai Li, Zhengming Ding, Kunpeng Li, Yulun Zhang, Yun Fu
Abstract	Person re-identification (re-ID) has recently been tremendously boosted due to the advancement of deep convolutional neural networks (CNN). The majority of deep re-ID methods focus on designing new CNN architectures, while less attention is paid on investigating the loss functions. Verification loss and identification loss are two types of losses widely used to train various deep re-ID models, both of which however have limitations. Verification loss guides the networks to generate feature embeddings of which the intra-class variance is decreased while the inter-class ones is enlarged. However, training networks with verification loss tends to be of slow convergence and unstable performance when the number of training samples is large. On the other hand, identification loss has good separating and scalable property. But its neglect to explicitly reduce the intra-class variance limits its performance on re-ID, because the same person may have significant appearance disparity across different camera views. To avoid the limitations of the two types of losses, we propose a new loss, called support neighbor (SN) loss. Rather than being derived from data sample pairs or triplets, SN loss is calculated based on the positive and negative support neighbor sets of each anchor sample, which contain more valuable contextual information and neighborhood structure that are beneficial for more stable performance. To ensure scalability and separability, a softmax-like function is formulated to push apart the positive and negative support sets. To reduce intra-class variance, the distance between the anchor’s nearest positive neighbor and furthest positive sample is penalized. Integrating SN loss on top of Resnet50, superior re-ID results to the state-of-the-art ones are obtained on several widely used datasets.
Tasks	Person Re-Identification
Published	2018-08-18
URL	http://arxiv.org/abs/1808.06030v1
PDF	http://arxiv.org/pdf/1808.06030v1.pdf
PWC	https://paperswithcode.com/paper/support-neighbor-loss-for-person-re
Repo	https://github.com/kailigo/SN_loss_for_reID
Framework	pytorch

fastMRI: An Open Dataset and Benchmarks for Accelerated MRI


Title	fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
Authors	Jure Zbontar, Florian Knoll, Anuroop Sriram, Tullie Murrell, Zhengnan Huang, Matthew J. Muckley, Aaron Defazio, Ruben Stern, Patricia Johnson, Mary Bruno, Marc Parente, Krzysztof J. Geras, Joe Katsnelson, Hersh Chandarana, Zizhao Zhang, Michal Drozdzal, Adriana Romero, Michael Rabbat, Pascal Vincent, Nafissa Yakubova, James Pinkerton, Duo Wang, Erich Owens, C. Lawrence Zitnick, Michael P. Recht, Daniel K. Sodickson, Yvonne W. Lui
Abstract	Accelerating Magnetic Resonance Imaging (MRI) by taking fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MRI possible in applications where it is currently prohibitively slow or expensive. We introduce the fastMRI dataset, a large-scale collection of both raw MR measurements and clinical MR images, that can be used for training and evaluation of machine-learning approaches to MR image reconstruction. By introducing standardized evaluation criteria and a freely-accessible dataset, our goal is to help the community make rapid advances in the state of the art for MR image reconstruction. We also provide a self-contained introduction to MRI for machine learning researchers with no medical imaging background.
Tasks	Image Reconstruction
Published	2018-11-21
URL	https://arxiv.org/abs/1811.08839v2
PDF	https://arxiv.org/pdf/1811.08839v2.pdf
PWC	https://paperswithcode.com/paper/fastmri-an-open-dataset-and-benchmarks-for
Repo	https://github.com/changheunoh/eternet_fastmri
Framework	none

How Many Random Seeds? Statistical Power Analysis in Deep Reinforcement Learning Experiments


Title	How Many Random Seeds? Statistical Power Analysis in Deep Reinforcement Learning Experiments
Authors	Cédric Colas, Olivier Sigaud, Pierre-Yves Oudeyer
Abstract	Consistently checking the statistical significance of experimental results is one of the mandatory methodological steps to address the so-called “reproducibility crisis” in deep reinforcement learning. In this tutorial paper, we explain how the number of random seeds relates to the probabilities of statistical errors. For both the t-test and the bootstrap confidence interval test, we recall theoretical guidelines to determine the number of random seeds one should use to provide a statistically significant comparison of the performance of two algorithms. Finally, we discuss the influence of deviations from the assumptions usually made by statistical tests. We show that they can lead to inaccurate evaluations of statistical errors and provide guidelines to counter these negative effects. We make our code available to perform the tests.
Tasks
Published	2018-06-21
URL	http://arxiv.org/abs/1806.08295v2
PDF	http://arxiv.org/pdf/1806.08295v2.pdf
PWC	https://paperswithcode.com/paper/how-many-random-seeds-statistical-power
Repo	https://github.com/flowersteam/rl-difference-testing
Framework	none

Preconditioner on Matrix Lie Group for SGD


Title	Preconditioner on Matrix Lie Group for SGD
Authors	Xi-Lin Li
Abstract	We study two types of preconditioners and preconditioned stochastic gradient descent (SGD) methods in a unified framework. We call the first one the Newton type due to its close relationship to the Newton method, and the second one the Fisher type as its preconditioner is closely related to the inverse of Fisher information matrix. Both preconditioners can be derived from one framework, and efficiently estimated on any matrix Lie groups designated by the user using natural or relative gradient descent minimizing certain preconditioner estimation criteria. Many existing preconditioners and methods, e.g., RMSProp, Adam, KFAC, equilibrated SGD, batch normalization, etc., are special cases of or closely related to either the Newton type or the Fisher type ones. Experimental results on relatively large scale machine learning problems are reported for performance study.
Tasks
Published	2018-09-26
URL	http://arxiv.org/abs/1809.10232v2
PDF	http://arxiv.org/pdf/1809.10232v2.pdf
PWC	https://paperswithcode.com/paper/preconditioner-on-matrix-lie-group-for-sgd
Repo	https://github.com/lixilinx/psgd_torch
Framework	pytorch

FingerVision Tactile Sensor Design and Slip Detection Using Convolutional LSTM Network


Title	FingerVision Tactile Sensor Design and Slip Detection Using Convolutional LSTM Network
Authors	Yazhan Zhang, Zicheng Kan, Yu Alexander Tse, Yang Yang, Michael Yu Wang
Abstract	Tactile sensing is essential to the human perception system, so as to robot. In this paper, we develop a novel optical-based tactile sensor “FingerVision” with effective signal processing algorithms. This sensor is composed of soft skin with embedded marker array bonded to rigid frame, and a web camera with a fisheye lens. While being excited with contact force, the camera tracks the movements of markers and deformation field is obtained. Compared to existing tactile sensors, our sensor features compact footprint, high resolution, and ease of fabrication. Besides, utilizing the deformation field estimation, we propose a slip classification framework based on convolution Long Short Term Memory (convolutional LSTM) networks. The data collection process takes advantage of the human sense of slip, during which human hand holds 12 daily objects, interacts with sensor skin and labels data with a slip or non-slip identity based on human feeling of slip. Our slip classification framework performs high accuracy of 97.62% on the test dataset. It is expected to be capable of enhancing the stability of robot grasping significantly, leading to better contact force control, finer object interaction and more active sensing manipulation.
Tasks
Published	2018-10-05
URL	http://arxiv.org/abs/1810.02653v1
PDF	http://arxiv.org/pdf/1810.02653v1.pdf
PWC	https://paperswithcode.com/paper/fingervision-tactile-sensor-design-and-slip
Repo	https://github.com/adamzhang129/fv_sensor
Framework	none

Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning


Title	Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning
Authors	Ke Yu, Chao Dong, Liang Lin, Chen Change Loy
Abstract	We investigate a novel approach for image restoration by reinforcement learning. Unlike existing studies that mostly train a single large network for a specialized task, we prepare a toolbox consisting of small-scale convolutional networks of different complexities and specialized in different tasks. Our method, RL-Restore, then learns a policy to select appropriate tools from the toolbox to progressively restore the quality of a corrupted image. We formulate a step-wise reward function proportional to how well the image is restored at each step to learn the action policy. We also devise a joint learning scheme to train the agent and tools for better performance in handling uncertainty. In comparison to conventional human-designed networks, RL-Restore is capable of restoring images corrupted with complex and unknown distortions in a more parameter-efficient manner using the dynamically formed toolchain.
Tasks	Image Restoration
Published	2018-04-10
URL	http://arxiv.org/abs/1804.03312v1
PDF	http://arxiv.org/pdf/1804.03312v1.pdf
PWC	https://paperswithcode.com/paper/crafting-a-toolchain-for-image-restoration-by
Repo	https://github.com/yuke93/RL-Restore
Framework	tf

Stochastic Distance Transform


Title	Stochastic Distance Transform
Authors	Johan Öfverstedt, Joakim Lindblad, Nataša Sladoje
Abstract	The distance transform (DT) and its many variations are ubiquitous tools for image processing and analysis. In many imaging scenarios, the images of interest are corrupted by noise. This has a strong negative impact on the accuracy of the DT, which is highly sensitive to spurious noise points. In this study, we consider images represented as discrete random sets and observe statistics of DT computed on such representations. We, thus, define a stochastic distance transform (SDT), which has an adjustable robustness to noise. Both a stochastic Monte Carlo method and a deterministic method for computing the SDT are proposed and compared. Through a series of empirical tests, we demonstrate that the SDT is effective not only in improving the accuracy of the computed distances in the presence of noise, but also in improving the performance of template matching and watershed segmentation of partially overlapping objects, which are examples of typical applications where DTs are utilized.
Tasks
Published	2018-10-18
URL	http://arxiv.org/abs/1810.08097v1
PDF	http://arxiv.org/pdf/1810.08097v1.pdf
PWC	https://paperswithcode.com/paper/stochastic-distance-transform
Repo	https://github.com/johanofverstedt/sdt
Framework	none

PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation


Title	PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation
Authors	Mingyang Jiang, Yiran Wu, Tianqi Zhao, Zelin Zhao, Cewu Lu
Abstract	Recently, 3D understanding research sheds light on extracting features from point cloud directly, which requires effective shape pattern description of point clouds. Inspired by the outstanding 2D shape descriptor SIFT, we design a module called PointSIFT that encodes information of different orientations and is adaptive to scale of shape. Specifically, an orientation-encoding unit is designed to describe eight crucial orientations, and multi-scale representation is achieved by stacking several orientation-encoding units. PointSIFT module can be integrated into various PointNet-based architecture to improve the representation ability. Extensive experiments show our PointSIFT-based framework outperforms state-of-the-art method on standard benchmark datasets. The code and trained model will be published accompanied by this paper.
Tasks	Semantic Segmentation
Published	2018-07-02
URL	http://arxiv.org/abs/1807.00652v2
PDF	http://arxiv.org/pdf/1807.00652v2.pdf
PWC	https://paperswithcode.com/paper/pointsift-a-sift-like-network-module-for-3d
Repo	https://github.com/583748495/psift
Framework	tf

Calibrating Deep Convolutional Gaussian Processes


Title	Calibrating Deep Convolutional Gaussian Processes
Authors	Gia-Lac Tran, Edwin V. Bonilla, John P. Cunningham, Pietro Michiardi, Maurizio Filippone
Abstract	The wide adoption of Convolutional Neural Networks (CNNs) in applications where decision-making under uncertainty is fundamental, has brought a great deal of attention to the ability of these models to accurately quantify the uncertainty in their predictions. Previous work on combining CNNs with Gaussian processes (GPs) has been developed under the assumption that the predictive probabilities of these models are well-calibrated. In this paper we show that, in fact, current combinations of CNNs and GPs are miscalibrated. We proposes a novel combination that considerably outperforms previous approaches on this aspect, while achieving state-of-the-art performance on image classification tasks.
Tasks	Decision Making, Decision Making Under Uncertainty, Gaussian Processes, Image Classification
Published	2018-05-26
URL	http://arxiv.org/abs/1805.10522v1
PDF	http://arxiv.org/pdf/1805.10522v1.pdf
PWC	https://paperswithcode.com/paper/calibrating-deep-convolutional-gaussian
Repo	https://github.com/GiaLacTRAN/convolutional_deep_gp_random_features
Framework	tf

Guided evolutionary strategies: Augmenting random search with surrogate gradients


Title	Guided evolutionary strategies: Augmenting random search with surrogate gradients
Authors	Niru Maheswaranathan, Luke Metz, George Tucker, Dami Choi, Jascha Sohl-Dickstein
Abstract	Many applications in machine learning require optimizing a function whose true gradient is unknown, but where surrogate gradient information (directions that may be correlated with, but not necessarily identical to, the true gradient) is available instead. This arises when an approximate gradient is easier to compute than the full gradient (e.g. in meta-learning or unrolled optimization), or when a true gradient is intractable and is replaced with a surrogate (e.g. in certain reinforcement learning applications, or when using synthetic gradients). We propose Guided Evolutionary Strategies, a method for optimally using surrogate gradient directions along with random search. We define a search distribution for evolutionary strategies that is elongated along a guiding subspace spanned by the surrogate gradients. This allows us to estimate a descent direction which can then be passed to a first-order optimizer. We analytically and numerically characterize the tradeoffs that result from tuning how strongly the search distribution is stretched along the guiding subspace, and we use this to derive a setting of the hyperparameters that works well across problems. Finally, we apply our method to example problems, demonstrating an improvement over both standard evolutionary strategies and first-order methods (that directly follow the surrogate gradient). We provide a demo of Guided ES at https://github.com/brain-research/guided-evolutionary-strategies
Tasks	Meta-Learning
Published	2018-06-26
URL	https://arxiv.org/abs/1806.10230v4
PDF	https://arxiv.org/pdf/1806.10230v4.pdf
PWC	https://paperswithcode.com/paper/guided-evolutionary-strategies-escaping-the
Repo	https://github.com/brain-research/guided-evolutionary-strategies
Framework	none

Entity Set Search of Scientific Literature: An Unsupervised Ranking Approach


Title	Entity Set Search of Scientific Literature: An Unsupervised Ranking Approach
Authors	Jiaming Shen, Jinfeng Xiao, Xinwei He, Jingbo Shang, Saurabh Sinha, Jiawei Han
Abstract	Literature search is critical for any scientific research. Different from Web or general domain search, a large portion of queries in scientific literature search are entity-set queries, that is, multiple entities of possibly different types. Entity-set queries reflect user’s need for finding documents that contain multiple entities and reveal inter-entity relationships and thus pose non-trivial challenges to existing search algorithms that model each entity separately. However, entity-set queries are usually sparse (i.e., not so repetitive), which makes ineffective many supervised ranking models that rely heavily on associated click history. To address these challenges, we introduce SetRank, an unsupervised ranking framework that models inter-entity relationships and captures entity type information. Furthermore, we develop a novel unsupervised model selection algorithm, based on the technique of weighted rank aggregation, to automatically choose the parameter settings in SetRank without resorting to a labeled validation set. We evaluate our proposed unsupervised approach using datasets from TREC Genomics Tracks and Semantic Scholar’s query log. The experiments demonstrate that SetRank significantly outperforms the baseline unsupervised models, especially on entity-set queries, and our model selection algorithm effectively chooses suitable parameter settings.
Tasks	Model Selection
Published	2018-04-29
URL	http://arxiv.org/abs/1804.10877v1
PDF	http://arxiv.org/pdf/1804.10877v1.pdf
PWC	https://paperswithcode.com/paper/180410877
Repo	https://github.com/mickeystroller/SetRank
Framework	none

Block-Structure Based Time-Series Models For Graph Sequences


Title	Block-Structure Based Time-Series Models For Graph Sequences
Authors	Mehrnaz Amjadi, Theja Tulabandhula
Abstract	Although the computational and statistical trade-off for modeling single graphs, for instance, using block models is relatively well understood, extending such results to sequences of graphs has proven to be difficult. In this work, we take a step in this direction by proposing two models for graph sequences that capture: (a) link persistence between nodes across time, and (b) community persistence of each node across time. In the first model, we assume that the latent community of each node does not change over time, and in the second model we relax this assumption suitably. For both of these proposed models, we provide statistically and computationally efficient inference algorithms, whose unique feature is that they leverage community detection methods that work on single graphs. We also provide experimental results validating the suitability of our models and methods on synthetic and real instances.
Tasks	Community Detection, Time Series
Published	2018-04-24
URL	http://arxiv.org/abs/1804.08796v2
PDF	http://arxiv.org/pdf/1804.08796v2.pdf
PWC	https://paperswithcode.com/paper/block-structure-based-time-series-models-for
Repo	https://github.com/thejat/dynamic-network-growth-models
Framework	none

Iterative Transformer Network for 3D Point Cloud


Title	Iterative Transformer Network for 3D Point Cloud
Authors	Wentao Yuan, David Held, Christoph Mertz, Martial Hebert
Abstract	3D point cloud is an efficient and flexible representation of 3D structures. Recently, neural networks operating on point clouds have shown superior performance on 3D understanding tasks such as shape classification and part segmentation. However, performance on such tasks is evaluated on complete shapes aligned in a canonical frame, while real world 3D data are partial and unaligned. A key challenge in learning from partial, unaligned point cloud data is to learn features that are invariant or equivariant with respect to geometric transformations. To address this challenge, we propose the Iterative Transformer Network (IT-Net), a network module that canonicalizes the pose of a partial object with a series of 3D rigid transformations predicted in an iterative fashion. We demonstrate the efficacy of IT-Net as an anytime pose estimator from partial point clouds without using complete object models. Further, we show that IT-Net achieves superior performance over alternative 3D transformer networks on various tasks, such as partial shape classification and object part segmentation.
Tasks
Published	2018-11-27
URL	https://arxiv.org/abs/1811.11209v2
PDF	https://arxiv.org/pdf/1811.11209v2.pdf
PWC	https://paperswithcode.com/paper/iterative-transformer-network-for-3d-point
Repo	https://github.com/wentaoyuan/it-net
Framework	tf

Parametric generation of conditional geological realizations using generative neural networks


Title	Parametric generation of conditional geological realizations using generative neural networks
Authors	Shing Chan, Ahmed H. Elsheikh
Abstract	Deep learning techniques are increasingly being considered for geological applications where – much like in computer vision – the challenges are characterized by high-dimensional spatial data dominated by multipoint statistics. In particular, a novel technique called generative adversarial networks has been recently studied for geological parametrization and synthesis, obtaining very impressive results that are at least qualitatively competitive with previous methods. The method obtains a neural network parametrization of the geology – so-called a generator – that is capable of reproducing very complex geological patterns with dimensionality reduction of several orders of magnitude. Subsequent works have addressed the conditioning task, i.e. using the generator to generate realizations honoring spatial observations (hard data). The current approaches, however, do not provide a parametrization of the conditional generation process. In this work, we propose a method to obtain a parametrization for direct generation of conditional realizations. The main idea is to simply extend the existing generator network by stacking a second inference network that learns to perform the conditioning. This inference network is a neural network trained to sample a posterior distribution derived using a Bayesian formulation of the conditioning task. The resulting extended neural network thus provides the conditional parametrization. Our method is assessed on a benchmark image of binary channelized subsurface, obtaining very promising results for a wide variety of conditioning configurations.
Tasks	Dimensionality Reduction
Published	2018-07-13
URL	https://arxiv.org/abs/1807.05207v2
PDF	https://arxiv.org/pdf/1807.05207v2.pdf
PWC	https://paperswithcode.com/paper/parametric-generation-of-conditional
Repo	https://github.com/chanshing/geocondition
Framework	pytorch