Paper Group ANR 552
SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation. Deep Convolutional Neural Network for 6-DOF Image Localization. Adversarial Influence Maximization. Learning from Binary Labels with Instance-Dependent Corruption. Mini-Batch Spectral Clustering. Robust imaging of hippocampal inner structure at 7T: in vivo acquisition protocol and me …
SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation
| Title | SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation |
| Authors | Li Yi, Hao Su, Xingwen Guo, Leonidas Guibas |
| Abstract | In this paper, we study the problem of semantic annotation on 3D models that are represented as shape graphs. A functional view is taken to represent localized information on graphs, so that annotations such as part segment or keypoint are nothing but 0-1 indicator vertex functions. Compared with images that are 2D grids, shape graphs are irregular and non-isomorphic data structures. To enable the prediction of vertex functions on them by convolutional neural networks, we resort to spectral CNN method that enables weight sharing by parameterizing kernels in the spectral domain spanned by graph laplacian eigenbases. Under this setting, our network, named SyncSpecCNN, strive to overcome two key challenges: how to share coefficients and conduct multi-scale analysis in different parts of the graph for a single shape, and how to share information across related but different shapes that may be represented by very different graphs. Towards these goals, we introduce a spectral parameterization of dilated convolutional kernels and a spectral transformer network. Experimentally we tested our SyncSpecCNN on various tasks, including 3D shape part segmentation and 3D keypoint prediction. State-of-the-art performance has been achieved on all benchmark datasets. |
| Tasks | 3D Part Segmentation |
| Published | 2016-12-02 |
| URL | http://arxiv.org/abs/1612.00606v1 |
| http://arxiv.org/pdf/1612.00606v1.pdf | |
| PWC | https://paperswithcode.com/paper/syncspeccnn-synchronized-spectral-cnn-for-3d |
| Repo | |
| Framework | |
Deep Convolutional Neural Network for 6-DOF Image Localization
| Title | Deep Convolutional Neural Network for 6-DOF Image Localization |
| Authors | Daoyuan Jia, Yongchi Su, Chunping Li |
| Abstract | We present an accurate and robust method for six degree of freedom image localization. There are two key-points of our method, 1. automatic immense photo synthesis and labeling from point cloud model and, 2. pose estimation with deep convolutional neural networks regression. Our model can directly regresses 6-DOF camera poses from images, accurately describing where and how it was captured. We achieved an accuracy within 1 meters and 1 degree on our out-door dataset, which covers about 2 acres on our school campus. |
| Tasks | Pose Estimation |
| Published | 2016-11-08 |
| URL | http://arxiv.org/abs/1611.02776v1 |
| http://arxiv.org/pdf/1611.02776v1.pdf | |
| PWC | https://paperswithcode.com/paper/deep-convolutional-neural-network-for-6-dof |
| Repo | |
| Framework | |
Adversarial Influence Maximization
| Title | Adversarial Influence Maximization |
| Authors | Justin Khim, Varun Jog, Po-Ling Loh |
| Abstract | We consider the problem of influence maximization in fixed networks for contagion models in an adversarial setting. The goal is to select an optimal set of nodes to seed the influence process, such that the number of influenced nodes at the conclusion of the campaign is as large as possible. We formulate the problem as a repeated game between a player and adversary, where the adversary specifies the edges along which the contagion may spread, and the player chooses sets of nodes to influence in an online fashion. We establish upper and lower bounds on the minimax pseudo-regret in both undirected and directed networks. |
| Tasks | |
| Published | 2016-11-01 |
| URL | http://arxiv.org/abs/1611.00350v2 |
| http://arxiv.org/pdf/1611.00350v2.pdf | |
| PWC | https://paperswithcode.com/paper/adversarial-influence-maximization |
| Repo | |
| Framework | |
Learning from Binary Labels with Instance-Dependent Corruption
| Title | Learning from Binary Labels with Instance-Dependent Corruption |
| Authors | Aditya Krishna Menon, Brendan van Rooyen, Nagarajan Natarajan |
| Abstract | Suppose we have a sample of instances paired with binary labels corrupted by arbitrary instance- and label-dependent noise. With sufficiently many such samples, can we optimally classify and rank instances with respect to the noise-free distribution? We provide a theoretical analysis of this question, with three main contributions. First, we prove that for instance-dependent noise, any algorithm that is consistent for classification on the noisy distribution is also consistent on the clean distribution. Second, we prove that for a broad class of instance- and label-dependent noise, a similar consistency result holds for the area under the ROC curve. Third, for the latter noise model, when the noise-free class-probability function belongs to the generalised linear model family, we show that the Isotron can efficiently and provably learn from the corrupted sample. |
| Tasks | |
| Published | 2016-05-03 |
| URL | http://arxiv.org/abs/1605.00751v2 |
| http://arxiv.org/pdf/1605.00751v2.pdf | |
| PWC | https://paperswithcode.com/paper/learning-from-binary-labels-with-instance |
| Repo | |
| Framework | |
Mini-Batch Spectral Clustering
| Title | Mini-Batch Spectral Clustering |
| Authors | Yufei Han, Maurizio Filippone |
| Abstract | The cost of computing the spectrum of Laplacian matrices hinders the application of spectral clustering to large data sets. While approximations recover computational tractability, they can potentially affect clustering performance. This paper proposes a practical approach to learn spectral clustering based on adaptive stochastic gradient optimization. Crucially, the proposed approach recovers the exact spectrum of Laplacian matrices in the limit of the iterations, and the cost of each iteration is linear in the number of samples. Extensive experimental validation on data sets with up to half a million samples demonstrate its scalability and its ability to outperform state-of-the-art approximate methods to learn spectral clustering for a given computational budget. |
| Tasks | |
| Published | 2016-07-07 |
| URL | http://arxiv.org/abs/1607.02024v2 |
| http://arxiv.org/pdf/1607.02024v2.pdf | |
| PWC | https://paperswithcode.com/paper/mini-batch-spectral-clustering |
| Repo | |
| Framework | |
Robust imaging of hippocampal inner structure at 7T: in vivo acquisition protocol and methodological choices
| Title | Robust imaging of hippocampal inner structure at 7T: in vivo acquisition protocol and methodological choices |
| Authors | Linda Marrakchi-Kacem, Alexandre Vignaud, Julien Sein, Johanne Germain, Thomas R Henry, Cyril Poupon, Lucie Hertz-Pannier, Stéphane Lehéricy, Olivier Colliot, Pierre-François Van de Moortele, Marie Chupin |
| Abstract | OBJECTIVE:Motion-robust multi-slab imaging of hippocampal inner structure in vivo at 7T.MATERIALS AND METHODS:Motion is a crucial issue for ultra-high resolution imaging, such as can be achieved with 7T MRI. An acquisition protocol was designed for imaging hippocampal inner structure at 7T. It relies on a compromise between anatomical details visibility and robustness to motion. In order to reduce acquisition time and motion artifacts, the full slab covering the hippocampus was split into separate slabs with lower acquisition time. A robust registration approach was implemented to combine the acquired slabs within a final 3D-consistent high-resolution slab covering the whole hippocampus. Evaluation was performed on 50 subjects overall, made of three groups of subjects acquired using three acquisition settings; it focused on three issues: visibility of hippocampal inner structure, robustness to motion artifacts and registration procedure performance.RESULTS:Overall, T2-weighted acquisitions with interleaved slabs proved robust. Multi-slab registration yielded high quality datasets in 96 % of the subjects, thus compatible with further analyses of hippocampal inner structure.CONCLUSION:Multi-slab acquisition and registration setting is efficient for reducing acquisition time and consequently motion artifacts for ultra-high resolution imaging of the inner structure of the hippocampus. |
| Tasks | |
| Published | 2016-05-09 |
| URL | http://arxiv.org/abs/1605.02559v1 |
| http://arxiv.org/pdf/1605.02559v1.pdf | |
| PWC | https://paperswithcode.com/paper/robust-imaging-of-hippocampal-inner-structure |
| Repo | |
| Framework | |
A Survey on Domain-Specific Languages for Machine Learning in Big Data
| Title | A Survey on Domain-Specific Languages for Machine Learning in Big Data |
| Authors | Ivens Portugal, Paulo Alencar, Donald Cowan |
| Abstract | The amount of data generated in the modern society is increasing rapidly. New problems and novel approaches of data capture, storage, analysis and visualization are responsible for the emergence of the Big Data research field. Machine Learning algorithms can be used in Big Data to make better and more accurate inferences. However, because of the challenges Big Data imposes, these algorithms need to be adapted and optimized to specific applications. One important decision made by software engineers is the choice of the language that is used in the implementation of these algorithms. Therefore, this literature survey identifies and describes domain-specific languages and frameworks used for Machine Learning in Big Data. By doing this, software engineers can then make more informed choices and beginners have an overview of the main languages used in this domain. |
| Tasks | |
| Published | 2016-02-24 |
| URL | http://arxiv.org/abs/1602.07637v2 |
| http://arxiv.org/pdf/1602.07637v2.pdf | |
| PWC | https://paperswithcode.com/paper/a-survey-on-domain-specific-languages-for |
| Repo | |
| Framework | |
Generalisation and Sharing in Triplet Convnets for Sketch based Visual Search
| Title | Generalisation and Sharing in Triplet Convnets for Sketch based Visual Search |
| Authors | Tu Bui, Leonardo Ribeiro, Moacir Ponti, John Collomosse |
| Abstract | We propose and evaluate several triplet CNN architectures for measuring the similarity between sketches and photographs, within the context of the sketch based image retrieval (SBIR) task. In contrast to recent fine-grained SBIR work, we study the ability of our networks to generalise across diverse object categories from limited training data, and explore in detail strategies for weight sharing, pre-processing, data augmentation and dimensionality reduction. We exceed the performance of pre-existing techniques on both the Flickr15k category level SBIR benchmark by $18%$, and the TU-Berlin SBIR benchmark by $\sim10 \mathcal{T}_b$, when trained on the 250 category TU-Berlin classification dataset augmented with 25k corresponding photographs harvested from the Internet. |
| Tasks | Data Augmentation, Dimensionality Reduction, Image Retrieval, Sketch-Based Image Retrieval |
| Published | 2016-11-16 |
| URL | http://arxiv.org/abs/1611.05301v1 |
| http://arxiv.org/pdf/1611.05301v1.pdf | |
| PWC | https://paperswithcode.com/paper/generalisation-and-sharing-in-triplet |
| Repo | |
| Framework | |
In Teacher We Trust: Learning Compressed Models for Pedestrian Detection
| Title | In Teacher We Trust: Learning Compressed Models for Pedestrian Detection |
| Authors | Jonathan Shen, Noranart Vesdapunt, Vishnu N. Boddeti, Kris M. Kitani |
| Abstract | Deep convolutional neural networks continue to advance the state-of-the-art in many domains as they grow bigger and more complex. It has been observed that many of the parameters of a large network are redundant, allowing for the possibility of learning a smaller network that mimics the outputs of the large network through a process called Knowledge Distillation. We show, however, that standard Knowledge Distillation is not effective for learning small models for the task of pedestrian detection. To improve this process, we introduce a higher-dimensional hint layer to increase information flow. We also estimate the variance in the outputs of the large network and propose a loss function to incorporate this uncertainty. Finally, we attempt to boost the complexity of the small network without increasing its size by using as input hand-designed features that have been demonstrated to be effective for pedestrian detection. We succeed in training a model that contains $400\times$ fewer parameters than the large network while outperforming AlexNet on the Caltech Pedestrian Dataset. |
| Tasks | Pedestrian Detection |
| Published | 2016-12-01 |
| URL | http://arxiv.org/abs/1612.00478v1 |
| http://arxiv.org/pdf/1612.00478v1.pdf | |
| PWC | https://paperswithcode.com/paper/in-teacher-we-trust-learning-compressed |
| Repo | |
| Framework | |
Unsupervised Learning with Imbalanced Data via Structure Consolidation Latent Variable Model
| Title | Unsupervised Learning with Imbalanced Data via Structure Consolidation Latent Variable Model |
| Authors | Fariba Yousefi, Zhenwen Dai, Carl Henrik Ek, Neil Lawrence |
| Abstract | Unsupervised learning on imbalanced data is challenging because, when given imbalanced data, current model is often dominated by the major category and ignores the categories with small amount of data. We develop a latent variable model that can cope with imbalanced data by dividing the latent space into a shared space and a private space. Based on Gaussian Process Latent Variable Models, we propose a new kernel formulation that enables the separation of latent space and derives an efficient variational inference method. The performance of our model is demonstrated with an imbalanced medical image dataset. |
| Tasks | Latent Variable Models |
| Published | 2016-06-30 |
| URL | http://arxiv.org/abs/1607.00067v1 |
| http://arxiv.org/pdf/1607.00067v1.pdf | |
| PWC | https://paperswithcode.com/paper/unsupervised-learning-with-imbalanced-data |
| Repo | |
| Framework | |
Multi-resolution Data Fusion for Super-Resolution Electron Microscopy
| Title | Multi-resolution Data Fusion for Super-Resolution Electron Microscopy |
| Authors | Suhas Sreehari, S. V. Venkatakrishnan, Katherine L. Bouman, Jeffrey P. Simmons, Lawrence F. Drummy, Charles A. Bouman |
| Abstract | Perhaps surprisingly, the total electron microscopy (EM) data collected to date is less than a cubic millimeter. Consequently, there is an enormous demand in the materials and biological sciences to image at greater speed and lower dosage, while maintaining resolution. Traditional EM imaging based on homogeneous raster-order scanning severely limits the volume of high-resolution data that can be collected, and presents a fundamental limitation to understanding physical processes such as material deformation, crack propagation, and pyrolysis. We introduce a novel multi-resolution data fusion (MDF) method for super-resolution computational EM. Our method combines innovative data acquisition with novel algorithmic techniques to dramatically improve the resolution/volume/speed trade-off. The key to our approach is to collect the entire sample at low resolution, while simultaneously collecting a small fraction of data at high resolution. The high-resolution measurements are then used to create a material-specific patch-library that is used within the “plug-and-play” framework to dramatically improve super-resolution of the low-resolution data. We present results using FEI electron microscope data that demonstrate super-resolution factors of 4x, 8x, and 16x, while substantially maintaining high image quality and reducing dosage. |
| Tasks | Super-Resolution |
| Published | 2016-11-28 |
| URL | http://arxiv.org/abs/1612.00874v1 |
| http://arxiv.org/pdf/1612.00874v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-resolution-data-fusion-for-super |
| Repo | |
| Framework | |
Latent-Class Hough Forests for 6 DoF Object Pose Estimation
| Title | Latent-Class Hough Forests for 6 DoF Object Pose Estimation |
| Authors | Rigas Kouskouridas, Alykhan Tejani, Andreas Doumanoglou, Danhang Tang, Tae-Kyun Kim |
| Abstract | In this paper we present Latent-Class Hough Forests, a method for object detection and 6 DoF pose estimation in heavily cluttered and occluded scenarios. We adapt a state of the art template matching feature into a scale-invariant patch descriptor and integrate it into a regression forest using a novel template-based split function. We train with positive samples only and we treat class distributions at the leaf nodes as latent variables. During testing we infer by iteratively updating these distributions, providing accurate estimation of background clutter and foreground occlusions and, thus, better detection rate. Furthermore, as a by-product, our Latent-Class Hough Forests can provide accurate occlusion aware segmentation masks, even in the multi-instance scenario. In addition to an existing public dataset, which contains only single-instance sequences with large amounts of clutter, we have collected two, more challenging, datasets for multiple-instance detection containing heavy 2D and 3D clutter as well as foreground occlusions. We provide extensive experiments on the various parameters of the framework such as patch size, number of trees and number of iterations to infer class distributions at test time. We also evaluate the Latent-Class Hough Forests on all datasets where we outperform state of the art methods. |
| Tasks | Object Detection, Pose Estimation |
| Published | 2016-02-03 |
| URL | http://arxiv.org/abs/1602.01464v1 |
| http://arxiv.org/pdf/1602.01464v1.pdf | |
| PWC | https://paperswithcode.com/paper/latent-class-hough-forests-for-6-dof-object |
| Repo | |
| Framework | |
Static and Dynamic Feature Selection in Morphosyntactic Analyzers
| Title | Static and Dynamic Feature Selection in Morphosyntactic Analyzers |
| Authors | Bernd Bohnet, Miguel Ballesteros, Ryan McDonald, Joakim Nivre |
| Abstract | We study the use of greedy feature selection methods for morphosyntactic tagging under a number of different conditions. We compare a static ordering of features to a dynamic ordering based on mutual information statistics, and we apply the techniques to standalone taggers as well as joint systems for tagging and parsing. Experiments on five languages show that feature selection can result in more compact models as well as higher accuracy under all conditions, but also that a dynamic ordering works better than a static ordering and that joint systems benefit more than standalone taggers. We also show that the same techniques can be used to select which morphosyntactic categories to predict in order to maximize syntactic accuracy in a joint system. Our final results represent a substantial improvement of the state of the art for several languages, while at the same time reducing both the number of features and the running time by up to 80% in some cases. |
| Tasks | Feature Selection |
| Published | 2016-03-21 |
| URL | http://arxiv.org/abs/1603.06503v1 |
| http://arxiv.org/pdf/1603.06503v1.pdf | |
| PWC | https://paperswithcode.com/paper/static-and-dynamic-feature-selection-in |
| Repo | |
| Framework | |
The Simulator: An Engine to Streamline Simulations
| Title | The Simulator: An Engine to Streamline Simulations |
| Authors | Jacob Bien |
| Abstract | The simulator is an R package that streamlines the process of performing simulations by creating a common infrastructure that can be easily used and reused across projects. Methodological statisticians routinely write simulations to compare their methods to preexisting ones. While developing ideas, there is a temptation to write “quick and dirty” simulations to try out ideas. This approach of rapid prototyping is useful but can sometimes backfire if bugs are introduced. Using the simulator allows one to remove the “dirty” without sacrificing the “quick.” Coding is quick because the statistician focuses exclusively on those aspects of the simulation that are specific to the particular paper being written. Code written with the simulator is succinct, highly readable, and easily shared with others. The modular nature of simulations written with the simulator promotes code reusability, which saves time and facilitates reproducibility. The syntax of the simulator leads to simulation code that is easily human-readable. Other benefits of using the simulator include the ability to “step in” to a simulation and change one aspect without having to rerun the entire simulation from scratch, the straightforward integration of parallel computing into simulations, and the ability to rapidly generate plots, tables, and reports with minimal effort. |
| Tasks | |
| Published | 2016-06-30 |
| URL | http://arxiv.org/abs/1607.00021v1 |
| http://arxiv.org/pdf/1607.00021v1.pdf | |
| PWC | https://paperswithcode.com/paper/the-simulator-an-engine-to-streamline |
| Repo | |
| Framework | |
Alternating Estimation for Structured High-Dimensional Multi-Response Models
| Title | Alternating Estimation for Structured High-Dimensional Multi-Response Models |
| Authors | Sheng Chen, Arindam Banerjee |
| Abstract | We consider learning high-dimensional multi-response linear models with structured parameters. By exploiting the noise correlations among responses, we propose an alternating estimation (AltEst) procedure to estimate the model parameters based on the generalized Dantzig selector. Under suitable sample size and resampling assumptions, we show that the error of the estimates generated by AltEst, with high probability, converges linearly to certain minimum achievable level, which can be tersely expressed by a few geometric measures, such as Gaussian width of sets related to the parameter structure. To the best of our knowledge, this is the first non-asymptotic statistical guarantee for such AltEst-type algorithm applied to estimation problem with general structures. |
| Tasks | |
| Published | 2016-06-29 |
| URL | http://arxiv.org/abs/1606.08957v1 |
| http://arxiv.org/pdf/1606.08957v1.pdf | |
| PWC | https://paperswithcode.com/paper/alternating-estimation-for-structured-high |
| Repo | |
| Framework | |
