May 7, 2019

2867 words 14 mins read

Paper Group AWR 77

Robust Lineage Reconstruction from High-Dimensional Single-Cell Data. Deep Learning in Multi-Layer Architectures of Dense Nuclei. Shape Constrained Tensor Decompositions using Sparse Representations in Over-Complete Libraries. Personalized Video Recommendation Using Rich Contents from Videos. Technical Report on the CleverHans v2.1.0 Adversarial Ex …

Robust Lineage Reconstruction from High-Dimensional Single-Cell Data


Title	Robust Lineage Reconstruction from High-Dimensional Single-Cell Data
Authors	Gregory Giecold, Eugenio Marco, Lorenzo Trippa, Guo-Cheng Yuan
Abstract	Single-cell gene expression data provide invaluable resources for systematic characterization of cellular hierarchy in multi-cellular organisms. However, cell lineage reconstruction is still often associated with significant uncertainty due to technological constraints. Such uncertainties have not been taken into account in current methods. We present ECLAIR, a novel computational method for the statistical inference of cell lineage relationships from single-cell gene expression data. ECLAIR uses an ensemble approach to improve the robustness of lineage predictions, and provides a quantitative estimate of the uncertainty of lineage branchings. We show that the application of ECLAIR to published datasets successfully reconstructs known lineage relationships and significantly improves the robustness of predictions. In conclusion, ECLAIR is a powerful bioinformatics tool for single-cell data analysis. It can be used for robust lineage reconstruction with quantitative estimate of prediction accuracy.
Tasks
Published	2016-01-12
URL	http://arxiv.org/abs/1601.02748v1
PDF	http://arxiv.org/pdf/1601.02748v1.pdf
PWC	https://paperswithcode.com/paper/robust-lineage-reconstruction-from-high
Repo	https://github.com/GGiecold/Cluster_Ensembles
Framework	none

Deep Learning in Multi-Layer Architectures of Dense Nuclei


Title	Deep Learning in Multi-Layer Architectures of Dense Nuclei
Authors	Yonghua Yin, Erol Gelenbe
Abstract	We assume that, within the dense clusters of neurons that can be found in nuclei, cells may interconnect via soma-to-soma interactions, in addition to conventional synaptic connections. We illustrate this idea with a multi-layer architecture (MLA) composed of multiple clusters of recurrent sub-networks of spiking Random Neural Networks (RNN) with dense soma-to-soma interactions, and use this RNN-MLA architecture for deep learning. The inputs to the clusters are first normalised by adjusting the external arrival rates of spikes to each cluster. Then we apply this architecture to learning from multi-channel datasets. Numerical results based on both images and sensor based data, show the value of this novel architecture for deep learning.
Tasks
Published	2016-09-22
URL	http://arxiv.org/abs/1609.07160v2
PDF	http://arxiv.org/pdf/1609.07160v2.pdf
PWC	https://paperswithcode.com/paper/deep-learning-in-multi-layer-architectures-of
Repo	https://github.com/yinyongh/DenseRandomNet
Framework	none

Shape Constrained Tensor Decompositions using Sparse Representations in Over-Complete Libraries


Title	Shape Constrained Tensor Decompositions using Sparse Representations in Over-Complete Libraries
Authors	Bethany Lusch, Eric C. Chi, J. Nathan Kutz
Abstract	We consider $N$-way data arrays and low-rank tensor factorizations where the time mode is coded as a sparse linear combination of temporal elements from an over-complete library. Our method, Shape Constrained Tensor Decomposition (SCTD) is based upon the CANDECOMP/PARAFAC (CP) decomposition which produces $r$-rank approximations of data tensors via outer products of vectors in each dimension of the data. By constraining the vector in the temporal dimension to known analytic forms which are selected from a large set of candidate functions, more readily interpretable decompositions are achieved and analytic time dependencies discovered. The SCTD method circumvents traditional {\em flattening} techniques where an $N$-way array is reshaped into a matrix in order to perform a singular value decomposition. A clear advantage of the SCTD algorithm is its ability to extract transient and intermittent phenomena which is often difficult for SVD-based methods. We motivate the SCTD method using several intuitively appealing results before applying it on a number of high-dimensional, real-world data sets in order to illustrate the efficiency of the algorithm in extracting interpretable spatio-temporal modes. With the rise of data-driven discovery methods, the decomposition proposed provides a viable technique for analyzing multitudes of data in a more comprehensible fashion.
Tasks
Published	2016-08-16
URL	http://arxiv.org/abs/1608.04674v1
PDF	http://arxiv.org/pdf/1608.04674v1.pdf
PWC	https://paperswithcode.com/paper/shape-constrained-tensor-decompositions-using
Repo	https://github.com/BethanyL/SCTD
Framework	none

Personalized Video Recommendation Using Rich Contents from Videos


Title	Personalized Video Recommendation Using Rich Contents from Videos
Authors	Xingzhong Du, Hongzhi Yin, Ling Chen, Yang Wang, Yi Yang, Xiaofang Zhou
Abstract	Video recommendation has become an essential way of helping people explore the massive videos and discover the ones that may be of interest to them. In the existing video recommender systems, the models make the recommendations based on the user-video interactions and single specific content features. When the specific content features are unavailable, the performance of the existing models will seriously deteriorate. Inspired by the fact that rich contents (e.g., text, audio, motion, and so on) exist in videos, in this paper, we explore how to use these rich contents to overcome the limitations caused by the unavailability of the specific ones. Specifically, we propose a novel general framework that incorporates arbitrary single content feature with user-video interactions, named as collaborative embedding regression (CER) model, to make effective video recommendation in both in-matrix and out-of-matrix scenarios. Our extensive experiments on two real-world large-scale datasets show that CER beats the existing recommender models with any single content feature and is more time efficient. In addition, we propose a priority-based late fusion (PRI) method to gain the benefit brought by the integrating the multiple content features. The corresponding experiment shows that PRI brings real performance improvement to the baseline and outperforms the existing fusion methods.
Tasks	Recommendation Systems
Published	2016-12-21
URL	http://arxiv.org/abs/1612.06935v6
PDF	http://arxiv.org/pdf/1612.06935v6.pdf
PWC	https://paperswithcode.com/paper/personalized-video-recommendation-using-rich
Repo	https://github.com/domainxz/top-k-rec
Framework	tf

Technical Report on the CleverHans v2.1.0 Adversarial Examples Library


Title	Technical Report on the CleverHans v2.1.0 Adversarial Examples Library
Authors	Nicolas Papernot, Fartash Faghri, Nicholas Carlini, Ian Goodfellow, Reuben Feinman, Alexey Kurakin, Cihang Xie, Yash Sharma, Tom Brown, Aurko Roy, Alexander Matyasko, Vahid Behzadan, Karen Hambardzumyan, Zhishuai Zhang, Yi-Lin Juang, Zhi Li, Ryan Sheatsley, Abhibhav Garg, Jonathan Uesato, Willi Gierke, Yinpeng Dong, David Berthelot, Paul Hendricks, Jonas Rauber, Rujun Long, Patrick McDaniel
Abstract	An adversarial example library for constructing attacks, building defenses, and benchmarking both
Tasks	Adversarial Attack, Adversarial Defense
Published	2016-10-03
URL	http://arxiv.org/abs/1610.00768v6
PDF	http://arxiv.org/pdf/1610.00768v6.pdf
PWC	https://paperswithcode.com/paper/technical-report-on-the-cleverhans-v210
Repo	https://github.com/AngusG/cleverhans-attacking-bnns
Framework	tf

Predicting Human Eye Fixations via an LSTM-based Saliency Attentive Model


Title	Predicting Human Eye Fixations via an LSTM-based Saliency Attentive Model
Authors	Marcella Cornia, Lorenzo Baraldi, Giuseppe Serra, Rita Cucchiara
Abstract	Data-driven saliency has recently gained a lot of attention thanks to the use of Convolutional Neural Networks for predicting gaze fixations. In this paper we go beyond standard approaches to saliency prediction, in which gaze maps are computed with a feed-forward network, and present a novel model which can predict accurate saliency maps by incorporating neural attentive mechanisms. The core of our solution is a Convolutional LSTM that focuses on the most salient regions of the input image to iteratively refine the predicted saliency map. Additionally, to tackle the center bias typical of human eye fixations, our model can learn a set of prior maps generated with Gaussian functions. We show, through an extensive evaluation, that the proposed architecture outperforms the current state of the art on public saliency prediction datasets. We further study the contribution of each key component to demonstrate their robustness on different scenarios.
Tasks	Saliency Prediction
Published	2016-11-29
URL	http://arxiv.org/abs/1611.09571v4
PDF	http://arxiv.org/pdf/1611.09571v4.pdf
PWC	https://paperswithcode.com/paper/predicting-human-eye-fixations-via-an-lstm
Repo	https://github.com/marcellacornia/sam
Framework	none

Convolutional Pose Machines


Title	Convolutional Pose Machines
Authors	Shih-En Wei, Varun Ramakrishna, Takeo Kanade, Yaser Sheikh
Abstract	Pose Machines provide a sequential prediction framework for learning rich implicit spatial models. In this work we show a systematic design for how convolutional networks can be incorporated into the pose machine framework for learning image features and image-dependent spatial models for the task of pose estimation. The contribution of this paper is to implicitly model long-range dependencies between variables in structured prediction tasks such as articulated pose estimation. We achieve this by designing a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference. Our approach addresses the characteristic difficulty of vanishing gradients during training by providing a natural learning objective function that enforces intermediate supervision, thereby replenishing back-propagated gradients and conditioning the learning procedure. We demonstrate state-of-the-art performance and outperform competing methods on standard benchmarks including the MPII, LSP, and FLIC datasets.
Tasks	3D Human Pose Estimation, Pose Estimation, Structured Prediction
Published	2016-01-30
URL	http://arxiv.org/abs/1602.00134v4
PDF	http://arxiv.org/pdf/1602.00134v4.pdf
PWC	https://paperswithcode.com/paper/convolutional-pose-machines
Repo	https://github.com/laobaiswag/openpose1
Framework	pytorch

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation


Title	PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Authors	Charles R. Qi, Hao Su, Kaichun Mo, Leonidas J. Guibas
Abstract	Point cloud is an important type of geometric data structure. Due to its irregular format, most researchers transform such data to regular 3D voxel grids or collections of images. This, however, renders data unnecessarily voluminous and causes issues. In this paper, we design a novel type of neural network that directly consumes point clouds and well respects the permutation invariance of points in the input. Our network, named PointNet, provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing. Though simple, PointNet is highly efficient and effective. Empirically, it shows strong performance on par or even better than state of the art. Theoretically, we provide analysis towards understanding of what the network has learnt and why the network is robust with respect to input perturbation and corruption.
Tasks	3D Part Segmentation, Object Classification, Scene Segmentation, Semantic Parsing, Semantic Segmentation, Skeleton Based Action Recognition
Published	2016-12-02
URL	http://arxiv.org/abs/1612.00593v2
PDF	http://arxiv.org/pdf/1612.00593v2.pdf
PWC	https://paperswithcode.com/paper/pointnet-deep-learning-on-point-sets-for-3d
Repo	https://github.com/YanWei123/Pointnet_encoder_Foldingnet_decoder_quantization
Framework	pytorch

CNN-RNN: A Unified Framework for Multi-label Image Classification


Title	CNN-RNN: A Unified Framework for Multi-label Image Classification
Authors	Jiang Wang, Yi Yang, Junhua Mao, Zhiheng Huang, Chang Huang, Wei Xu
Abstract	While deep convolutional neural networks (CNNs) have shown a great success in single-label image classification, it is important to note that real world images generally contain multiple labels, which could correspond to different objects, scenes, actions and attributes in an image. Traditional approaches to multi-label image classification learn independent classifiers for each category and employ ranking or thresholding on the classification results. These techniques, although working well, fail to explicitly exploit the label dependencies in an image. In this paper, we utilize recurrent neural networks (RNNs) to address this problem. Combined with CNNs, the proposed CNN-RNN framework learns a joint image-label embedding to characterize the semantic label dependency as well as the image-label relevance, and it can be trained end-to-end from scratch to integrate both information in a unified framework. Experimental results on public benchmark datasets demonstrate that the proposed architecture achieves better performance than the state-of-the-art multi-label classification model
Tasks	Image Classification, Multi-Label Classification
Published	2016-04-15
URL	http://arxiv.org/abs/1604.04573v1
PDF	http://arxiv.org/pdf/1604.04573v1.pdf
PWC	https://paperswithcode.com/paper/cnn-rnn-a-unified-framework-for-multi-label
Repo	https://github.com/Epiphqny/Multiple-instance-learning
Framework	pytorch

Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering


Title	Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering
Authors	Chong You, Chun-Guang Li, Daniel P. Robinson, Rene Vidal
Abstract	State-of-the-art subspace clustering methods are based on expressing each data point as a linear combination of other data points while regularizing the matrix of coefficients with $\ell_1$, $\ell_2$ or nuclear norms. $\ell_1$ regularization is guaranteed to give a subspace-preserving affinity (i.e., there are no connections between points from different subspaces) under broad theoretical conditions, but the clusters may not be connected. $\ell_2$ and nuclear norm regularization often improve connectivity, but give a subspace-preserving affinity only for independent subspaces. Mixed $\ell_1$, $\ell_2$ and nuclear norm regularizations offer a balance between the subspace-preserving and connectedness properties, but this comes at the cost of increased computational complexity. This paper studies the geometry of the elastic net regularizer (a mixture of the $\ell_1$ and $\ell_2$ norms) and uses it to derive a provably correct and scalable active set method for finding the optimal coefficients. Our geometric analysis also provides a theoretical justification and a geometric interpretation for the balance between the connectedness (due to $\ell_2$ regularization) and subspace-preserving (due to $\ell_1$ regularization) properties for elastic net subspace clustering. Our experiments show that the proposed active set method not only achieves state-of-the-art clustering performance, but also efficiently handles large-scale datasets.
Tasks
Published	2016-05-09
URL	http://arxiv.org/abs/1605.02633v1
PDF	http://arxiv.org/pdf/1605.02633v1.pdf
PWC	https://paperswithcode.com/paper/oracle-based-active-set-algorithm-for
Repo	https://github.com/ChongYou/subspace-clustering
Framework	none

Structured Receptive Fields in CNNs


Title	Structured Receptive Fields in CNNs
Authors	Jörn-Henrik Jacobsen, Jan van Gemert, Zhongyu Lou, Arnold W. M. Smeulders
Abstract	Learning powerful feature representations with CNNs is hard when training data are limited. Pre-training is one way to overcome this, but it requires large datasets sufficiently similar to the target domain. Another option is to design priors into the model, which can range from tuned hyperparameters to fully engineered representations like Scattering Networks. We combine these ideas into structured receptive field networks, a model which has a fixed filter basis and yet retains the flexibility of CNNs. This flexibility is achieved by expressing receptive fields in CNNs as a weighted sum over a fixed basis which is similar in spirit to Scattering Networks. The key difference is that we learn arbitrary effective filter sets from the basis rather than modeling the filters. This approach explicitly connects classical multiscale image analysis with general CNNs. With structured receptive field networks, we improve considerably over unstructured CNNs for small and medium dataset scenarios as well as over Scattering for large datasets. We validate our findings on ILSVRC2012, Cifar-10, Cifar-100 and MNIST. As a realistic small dataset example, we show state-of-the-art classification results on popular 3D MRI brain-disease datasets where pre-training is difficult due to a lack of large public datasets in a similar domain.
Tasks
Published	2016-05-10
URL	http://arxiv.org/abs/1605.02971v2
PDF	http://arxiv.org/pdf/1605.02971v2.pdf
PWC	https://paperswithcode.com/paper/structured-receptive-fields-in-cnns
Repo	https://github.com/nkarantzas/quick_draw
Framework	pytorch

Bidirectional LSTM-CRF for Clinical Concept Extraction


Title	Bidirectional LSTM-CRF for Clinical Concept Extraction
Authors	Raghavendra Chalapathy, Ehsan Zare Borzeshi, Massimo Piccardi
Abstract	Extraction of concepts present in patient clinical records is an essential step in clinical research. The 2010 i2b2/VA Workshop on Natural Language Processing Challenges for clinical records presented concept extraction (CE) task, with aim to identify concepts (such as treatments, tests, problems) and classify them into predefined categories. State-of-the-art CE approaches heavily rely on hand crafted features and domain specific resources which are hard to collect and tune. For this reason, this paper employs bidirectional LSTM with CRF decoding initialized with general purpose off-the-shelf word embeddings for CE. The experimental results achieved on 2010 i2b2/VA reference standard corpora using bidirectional LSTM CRF ranks closely with top ranked systems.
Tasks	Clinical Concept Extraction, Word Embeddings
Published	2016-10-19
URL	http://arxiv.org/abs/1610.05858v1
PDF	http://arxiv.org/pdf/1610.05858v1.pdf
PWC	https://paperswithcode.com/paper/bidirectional-lstm-crf-for-clinical-concept-1
Repo	https://github.com/raghavchalapathy/Bidirectional-LSTM-CRF-for-Clinical-Concept-Extraction
Framework	none

Deep Saliency with Encoded Low level Distance Map and High Level Features


Title	Deep Saliency with Encoded Low level Distance Map and High Level Features
Authors	Gayoung Lee, Yu-Wing Tai, Junmo Kim
Abstract	Recent advances in saliency detection have utilized deep learning to obtain high level features to detect salient regions in a scene. These advances have demonstrated superior results over previous works that utilize hand-crafted low level features for saliency detection. In this paper, we demonstrate that hand-crafted features can provide complementary information to enhance performance of saliency detection that utilizes only high level features. Our method utilizes both high level and low level features for saliency detection under a unified deep learning framework. The high level features are extracted using the VGG-net, and the low level features are compared with other parts of an image to form a low level distance map. The low level distance map is then encoded using a convolutional neural network(CNN) with multiple 1X1 convolutional and ReLU layers. We concatenate the encoded low level distance map and the high level features, and connect them to a fully connected neural network classifier to evaluate the saliency of a query region. Our experiments show that our method can further improve the performance of state-of-the-art deep learning-based saliency detection methods.
Tasks	Saliency Detection
Published	2016-04-19
URL	http://arxiv.org/abs/1604.05495v1
PDF	http://arxiv.org/pdf/1604.05495v1.pdf
PWC	https://paperswithcode.com/paper/deep-saliency-with-encoded-low-level-distance
Repo	https://github.com/gylee1103/SaliencyELD
Framework	none

Generalization Properties of Learning with Random Features


Title	Generalization Properties of Learning with Random Features
Authors	Alessandro Rudi, Lorenzo Rosasco
Abstract	We study the generalization properties of ridge regression with random features in the statistical learning framework. We show for the first time that $O(1/\sqrt{n})$ learning bounds can be achieved with only $O(\sqrt{n}\log n)$ random features rather than $O({n})$ as suggested by previous results. Further, we prove faster learning rates and show that they might require more random features, unless they are sampled according to a possibly problem dependent distribution. Our results shed light on the statistical computational trade-offs in large scale kernelized learning, showing the potential effectiveness of random features in reducing the computational complexity while keeping optimal generalization properties.
Tasks
Published	2016-02-14
URL	http://arxiv.org/abs/1602.04474v4
PDF	http://arxiv.org/pdf/1602.04474v4.pdf
PWC	https://paperswithcode.com/paper/generalization-properties-of-learning-with
Repo	https://github.com/dnbaker/frp
Framework	none

Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations


Title	Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations
Authors	Aaron Schein, Mingyuan Zhou, David M. Blei, Hanna Wallach
Abstract	We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country–country interaction event data. These data consist of interaction events of the form “country $i$ took action $a$ toward country $j$ at time $t$.” BPTD discovers overlapping country–community memberships, including the number of latent communities. In addition, it discovers directed community–community interaction networks that are specific to “topics” of action types and temporal “regimes.” We show that BPTD yields an efficient MCMC inference algorithm and achieves better predictive performance than related models. We also demonstrate that it discovers interpretable latent structure that agrees with our knowledge of international relations.
Tasks
Published	2016-06-06
URL	http://arxiv.org/abs/1606.01855v1
PDF	http://arxiv.org/pdf/1606.01855v1.pdf
PWC	https://paperswithcode.com/paper/bayesian-poisson-tucker-decomposition-for
Repo	https://github.com/aschein/bptd
Framework	none