Paper Group ANR 66
Scalable Adaptive Stochastic Optimization Using Random Projections. A-expansion for multiple “hedgehog” shapes. Compression and the origins of Zipf’s law for word frequencies. Sub-Linear Privacy-Preserving Near-Neighbor Search. DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression. Image Credibility Analysis with Effect …
Scalable Adaptive Stochastic Optimization Using Random Projections
| Title | Scalable Adaptive Stochastic Optimization Using Random Projections |
| Authors | Gabriel Krummenacher, Brian McWilliams, Yannic Kilcher, Joachim M. Buhmann, Nicolai Meinshausen |
| Abstract | Adaptive stochastic gradient methods such as AdaGrad have gained popularity in particular for training deep neural networks. The most commonly used and studied variant maintains a diagonal matrix approximation to second order information by accumulating past gradients which are used to tune the step size adaptively. In certain situations the full-matrix variant of AdaGrad is expected to attain better performance, however in high dimensions it is computationally impractical. We present Ada-LR and RadaGrad two computationally efficient approximations to full-matrix AdaGrad based on randomized dimensionality reduction. They are able to capture dependencies between features and achieve similar performance to full-matrix AdaGrad but at a much smaller computational cost. We show that the regret of Ada-LR is close to the regret of full-matrix AdaGrad which can have an up-to exponentially smaller dependence on the dimension than the diagonal variant. Empirically, we show that Ada-LR and RadaGrad perform similarly to full-matrix AdaGrad. On the task of training convolutional neural networks as well as recurrent neural networks, RadaGrad achieves faster convergence than diagonal AdaGrad. |
| Tasks | Dimensionality Reduction, Stochastic Optimization |
| Published | 2016-11-21 |
| URL | http://arxiv.org/abs/1611.06652v1 |
| http://arxiv.org/pdf/1611.06652v1.pdf | |
| PWC | https://paperswithcode.com/paper/scalable-adaptive-stochastic-optimization |
| Repo | |
| Framework | |
A-expansion for multiple “hedgehog” shapes
| Title | A-expansion for multiple “hedgehog” shapes |
| Authors | Hossam Isack, Yuri Boykov, Olga Veksler |
| Abstract | Overlapping colors and cluttered or weak edges are common segmentation problems requiring additional regularization. For example, star-convexity is popular for interactive single object segmentation due to simplicity and amenability to exact graph cut optimization. This paper proposes an approach to multiobject segmentation where objects could be restricted to separate “hedgehog” shapes. We show that a-expansion moves are submodular for our multi-shape constraints. Each “hedgehog” shape has its surface normals constrained by some vector field, e.g. gradients of a distance transform for user scribbles. Tight constraint give an extreme case of a shape prior enforcing skeleton consistency with the scribbles. Wider cones of allowed normals gives more relaxed hedgehog shapes. A single click and +/-90 degrees normal orientation constraints reduce our hedgehog prior to star-convexity. If all hedgehogs come from single clicks then our approach defines multi-star prior. Our general method has significantly more applications than standard one-star segmentation. For example, in medical data we can separate multiple non-star organs with similar appearances and weak or noisy edges. |
| Tasks | Semantic Segmentation |
| Published | 2016-02-02 |
| URL | http://arxiv.org/abs/1602.01006v1 |
| http://arxiv.org/pdf/1602.01006v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-expansion-for-multiple-hedgehog-shapes |
| Repo | |
| Framework | |
Compression and the origins of Zipf’s law for word frequencies
| Title | Compression and the origins of Zipf’s law for word frequencies |
| Authors | Ramon Ferrer-i-Cancho |
| Abstract | Here we sketch a new derivation of Zipf’s law for word frequencies based on optimal coding. The structure of the derivation is reminiscent of Mandelbrot’s random typing model but it has multiple advantages over random typing: (1) it starts from realistic cognitive pressures (2) it does not require fine tuning of parameters and (3) it sheds light on the origins of other statistical laws of language and thus can lead to a compact theory of linguistic laws. Our findings suggest that the recurrence of Zipf’s law in human languages could originate from pressure for easy and fast communication. |
| Tasks | |
| Published | 2016-05-04 |
| URL | http://arxiv.org/abs/1605.01326v2 |
| http://arxiv.org/pdf/1605.01326v2.pdf | |
| PWC | https://paperswithcode.com/paper/compression-and-the-origins-of-zipfs-law-for |
| Repo | |
| Framework | |
Sub-Linear Privacy-Preserving Near-Neighbor Search
| Title | Sub-Linear Privacy-Preserving Near-Neighbor Search |
| Authors | M. Sadegh Riazi, Beidi Chen, Anshumali Shrivastava, Dan Wallach, Farinaz Koushanfar |
| Abstract | In Near-Neighbor Search (NNS), a new client queries a database (held by a server) for the most similar data (near-neighbors) given a certain similarity metric. The Privacy-Preserving variant (PP-NNS) requires that neither server nor the client shall learn information about the other party’s data except what can be inferred from the outcome of NNS. The overwhelming growth in the size of current datasets and the lack of a truly secure server in the online world render the existing solutions impractical; either due to their high computational requirements or non-realistic assumptions which potentially compromise privacy. PP-NNS having query time {\it sub-linear} in the size of the database has been suggested as an open research direction by Li et al. (CCSW’15). In this paper, we provide the first such algorithm, called Secure Locality Sensitive Indexing (SLSI) which has a sub-linear query time and the ability to handle honest-but-curious parties. At the heart of our proposal lies a secure binary embedding scheme generated from a novel probabilistic transformation over locality sensitive hashing family. We provide information theoretic bound for the privacy guarantees and support our theoretical claims using substantial empirical evidence on real-world datasets. |
| Tasks | |
| Published | 2016-12-06 |
| URL | https://arxiv.org/abs/1612.01835v4 |
| https://arxiv.org/pdf/1612.01835v4.pdf | |
| PWC | https://paperswithcode.com/paper/sub-linear-privacy-preserving-near-neighbor |
| Repo | |
| Framework | |
DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression
| Title | DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression |
| Authors | Jovana Mitrovic, Dino Sejdinovic, Yee Whye Teh |
| Abstract | Performing exact posterior inference in complex generative models is often difficult or impossible due to an expensive to evaluate or intractable likelihood function. Approximate Bayesian computation (ABC) is an inference framework that constructs an approximation to the true likelihood based on the similarity between the observed and simulated data as measured by a predefined set of summary statistics. Although the choice of appropriate problem-specific summary statistics crucially influences the quality of the likelihood approximation and hence also the quality of the posterior sample in ABC, there are only few principled general-purpose approaches to the selection or construction of such summary statistics. In this paper, we develop a novel framework for this task using kernel-based distribution regression. We model the functional relationship between data distributions and the optimal choice (with respect to a loss function) of summary statistics using kernel-based distribution regression. We show that our approach can be implemented in a computationally and statistically efficient way using the random Fourier features framework for large-scale kernel learning. In addition to that, our framework shows superior performance when compared to related methods on toy and real-world problems. |
| Tasks | |
| Published | 2016-02-15 |
| URL | http://arxiv.org/abs/1602.04805v1 |
| http://arxiv.org/pdf/1602.04805v1.pdf | |
| PWC | https://paperswithcode.com/paper/dr-abc-approximate-bayesian-computation-with |
| Repo | |
| Framework | |
Image Credibility Analysis with Effective Domain Transferred Deep Networks
| Title | Image Credibility Analysis with Effective Domain Transferred Deep Networks |
| Authors | Zhiwei Jin, Juan Cao, Jiebo Luo, Yongdong Zhang |
| Abstract | Numerous fake images spread on social media today and can severely jeopardize the credibility of online content to public. In this paper, we employ deep networks to learn distinct fake image related features. In contrast to authentic images, fake images tend to be eye-catching and visually striking. Compared with traditional visual recognition tasks, it is extremely challenging to understand these psychologically triggered visual patterns in fake images. Traditional general image classification datasets, such as ImageNet set, are designed for feature learning at the object level but are not suitable for learning the hyper-features that would be required by image credibility analysis. In order to overcome the scarcity of training samples of fake images, we first construct a large-scale auxiliary dataset indirectly related to this task. This auxiliary dataset contains 0.6 million weakly-labeled fake and real images collected automatically from social media. Through an AdaBoost-like transfer learning algorithm, we train a CNN model with a few instances in the target training set and 0.6 million images in the collected auxiliary set. This learning algorithm is able to leverage knowledge from the auxiliary set and gradually transfer it to the target task. Experiments on a real-world testing set show that our proposed domain transferred CNN model outperforms several competing baselines. It obtains superiror results over transfer learning methods based on the general ImageNet set. Moreover, case studies show that our proposed method reveals some interesting patterns for distinguishing fake and authentic images. |
| Tasks | Image Classification, Transfer Learning |
| Published | 2016-11-16 |
| URL | http://arxiv.org/abs/1611.05328v1 |
| http://arxiv.org/pdf/1611.05328v1.pdf | |
| PWC | https://paperswithcode.com/paper/image-credibility-analysis-with-effective |
| Repo | |
| Framework | |
Head Pose Estimation of Occluded Faces using Regularized Regression
| Title | Head Pose Estimation of Occluded Faces using Regularized Regression |
| Authors | Amit Kumar, Rishabh Bindal, Soumya Indela, Michael Rotkowitz |
| Abstract | This paper presents regression methods for estimation of head pose from occluded 2-D face images. The process primarily involves reconstructing a face from its occluded image, followed by classification. Typical methods for reconstruction assume that the pixel errors of the occluded regions are independent. However, such an assumption is not true in the case of occlusion, because of its inherent contiguous nature. Hence, we use nuclear norm as a metric that can describe well the structure of the error. We also use LASSO Regression based l1 - regularization to improve reconstruction. Next, we implement Nuclear Norm Regularized Regression (NR), and also our proposed method, for reconstruction and subsequent classification. Finally, we compare the performance of the methods in terms of accuracy of head pose estimation of occluded faces. |
| Tasks | Head Pose Estimation, Pose Estimation |
| Published | 2016-02-02 |
| URL | http://arxiv.org/abs/1602.00997v1 |
| http://arxiv.org/pdf/1602.00997v1.pdf | |
| PWC | https://paperswithcode.com/paper/head-pose-estimation-of-occluded-faces-using |
| Repo | |
| Framework | |
Recurrent Image Captioner: Describing Images with Spatial-Invariant Transformation and Attention Filtering
| Title | Recurrent Image Captioner: Describing Images with Spatial-Invariant Transformation and Attention Filtering |
| Authors | Hao Liu, Yang Yang, Fumin Shen, Lixin Duan, Heng Tao Shen |
| Abstract | Along with the prosperity of recurrent neural network in modelling sequential data and the power of attention mechanism in automatically identify salient information, image captioning, a.k.a., image description, has been remarkably advanced in recent years. Nonetheless, most existing paradigms may suffer from the deficiency of invariance to images with different scaling, rotation, etc.; and effective integration of standalone attention to form a holistic end-to-end system. In this paper, we propose a novel image captioning architecture, termed Recurrent Image Captioner (\textbf{RIC}), which allows visual encoder and language decoder to coherently cooperate in a recurrent manner. Specifically, we first equip CNN-based visual encoder with a differentiable layer to enable spatially invariant transformation of visual signals. Moreover, we deploy an attention filter module (differentiable) between encoder and decoder to dynamically determine salient visual parts. We also employ bidirectional LSTM to preprocess sentences for generating better textual representations. Besides, we propose to exploit variational inference to optimize the whole architecture. Extensive experimental results on three benchmark datasets (i.e., Flickr8k, Flickr30k and MS COCO) demonstrate the superiority of our proposed architecture as compared to most of the state-of-the-art methods. |
| Tasks | Image Captioning |
| Published | 2016-12-15 |
| URL | http://arxiv.org/abs/1612.04949v1 |
| http://arxiv.org/pdf/1612.04949v1.pdf | |
| PWC | https://paperswithcode.com/paper/recurrent-image-captioner-describing-images |
| Repo | |
| Framework | |
Toward Efficient Task Assignment and Motion Planning for Large Scale Underwater Mission
| Title | Toward Efficient Task Assignment and Motion Planning for Large Scale Underwater Mission |
| Authors | Somaiyeh Mahmoud Zadeh, David MW Powers, Karl Sammut, Amirmehdi Yazdani |
| Abstract | An Autonomous Underwater Vehicle (AUV) needs to acquire a certain degree of autonomy for any particular underwater mission to fulfill the mission objectives successfully and ensure its safety in all stages of the mission in a large scale operating filed. In this paper, a novel combinatorial conflict-free-task assignment strategy consisting an interactive engagement of a local path planner and an adaptive global route planner, is introduced. The method is established upon the heuristic search potency of the Particle Swarm Optimisation (PSO) algorithm to address the discrete nature of routing-task assignment approach and the complexity of NP-hard path planning problem. The proposed hybrid method is highly efficient for having a reactive guidance framework that guarantees successful completion of missions specifically in cluttered environments. To examine the performance of the method in a context of mission productivity, mission time management and vehicle safety, a series of simulation studies are undertaken. The results of simulations declare that the proposed method is reliable and robust, particularly in dealing with uncertainties, and it can significantly enhance the level of vehicle’s autonomy by relying on its reactive nature and capability of providing fast feasible solutions. |
| Tasks | Motion Planning |
| Published | 2016-04-17 |
| URL | http://arxiv.org/abs/1604.04854v3 |
| http://arxiv.org/pdf/1604.04854v3.pdf | |
| PWC | https://paperswithcode.com/paper/toward-efficient-task-assignment-and-motion |
| Repo | |
| Framework | |
Active and Continuous Exploration with Deep Neural Networks and Expected Model Output Changes
| Title | Active and Continuous Exploration with Deep Neural Networks and Expected Model Output Changes |
| Authors | Christoph Käding, Erik Rodner, Alexander Freytag, Joachim Denzler |
| Abstract | The demands on visual recognition systems do not end with the complexity offered by current large-scale image datasets, such as ImageNet. In consequence, we need curious and continuously learning algorithms that actively acquire knowledge about semantic concepts which are present in available unlabeled data. As a step towards this goal, we show how to perform continuous active learning and exploration, where an algorithm actively selects relevant batches of unlabeled examples for annotation. These examples could either belong to already known or to yet undiscovered classes. Our algorithm is based on a new generalization of the Expected Model Output Change principle for deep architectures and is especially tailored to deep neural networks. Furthermore, we show easy-to-implement approximations that yield efficient techniques for active selection. Empirical experiments show that our method outperforms currently used heuristics. |
| Tasks | Active Learning |
| Published | 2016-12-19 |
| URL | http://arxiv.org/abs/1612.06129v1 |
| http://arxiv.org/pdf/1612.06129v1.pdf | |
| PWC | https://paperswithcode.com/paper/active-and-continuous-exploration-with-deep |
| Repo | |
| Framework | |
A Statistical Model for Stroke Outcome Prediction and Treatment Planning
| Title | A Statistical Model for Stroke Outcome Prediction and Treatment Planning |
| Authors | Abhishek Sengupta, Vaibhav Rajan, Sakyajit Bhattacharya, G R K Sarma |
| Abstract | Stroke is a major cause of mortality and long–term disability in the world. Predictive outcome models in stroke are valuable for personalized treatment, rehabilitation planning and in controlled clinical trials. In this paper we design a new model to predict outcome in the short-term, the putative therapeutic window for several treatments. Our regression-based model has a parametric form that is designed to address many challenges common in medical datasets like highly correlated variables and class imbalance. Empirically our model outperforms the best–known previous models in predicting short–term outcomes and in inferring the most effective treatments that improve outcome. |
| Tasks | |
| Published | 2016-02-22 |
| URL | http://arxiv.org/abs/1602.07280v1 |
| http://arxiv.org/pdf/1602.07280v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-statistical-model-for-stroke-outcome |
| Repo | |
| Framework | |
With Whom Do I Interact? Detecting Social Interactions in Egocentric Photo-streams
| Title | With Whom Do I Interact? Detecting Social Interactions in Egocentric Photo-streams |
| Authors | Maedeh Aghaei, Mariella Dimiccoli, Petia Radeva |
| Abstract | Given a user wearing a low frame rate wearable camera during a day, this work aims to automatically detect the moments when the user gets engaged into a social interaction solely by reviewing the automatically captured photos by the worn camera. The proposed method, inspired by the sociological concept of F-formation, exploits distance and orientation of the appearing individuals -with respect to the user- in the scene from a bird-view perspective. As a result, the interaction pattern over the sequence can be understood as a two-dimensional time series that corresponds to the temporal evolution of the distance and orientation features over time. A Long-Short Term Memory-based Recurrent Neural Network is then trained to classify each time series. Experimental evaluation over a dataset of 30.000 images has shown promising results on the proposed method for social interaction detection in egocentric photo-streams. |
| Tasks | Time Series |
| Published | 2016-05-13 |
| URL | http://arxiv.org/abs/1605.04129v2 |
| http://arxiv.org/pdf/1605.04129v2.pdf | |
| PWC | https://paperswithcode.com/paper/with-whom-do-i-interact-detecting-social |
| Repo | |
| Framework | |
End-to-End Radio Traffic Sequence Recognition with Deep Recurrent Neural Networks
| Title | End-to-End Radio Traffic Sequence Recognition with Deep Recurrent Neural Networks |
| Authors | Timothy J. O’Shea, Seth Hitefield, Johnathan Corgan |
| Abstract | We investigate sequence machine learning techniques on raw radio signal time-series data. By applying deep recurrent neural networks we learn to discriminate between several application layer traffic types on top of a constant envelope modulation without using an expert demodulation algorithm. We show that complex protocol sequences can be learned and used for both classification and generation tasks using this approach. |
| Tasks | Time Series |
| Published | 2016-10-03 |
| URL | http://arxiv.org/abs/1610.00564v1 |
| http://arxiv.org/pdf/1610.00564v1.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-radio-traffic-sequence-recognition |
| Repo | |
| Framework | |
DeepBinaryMask: Learning a Binary Mask for Video Compressive Sensing
| Title | DeepBinaryMask: Learning a Binary Mask for Video Compressive Sensing |
| Authors | Michael Iliadis, Leonidas Spinoulas, Aggelos K. Katsaggelos |
| Abstract | In this paper, we propose a novel encoder-decoder neural network model referred to as DeepBinaryMask for video compressive sensing. In video compressive sensing one frame is acquired using a set of coded masks (sensing matrix) from which a number of video frames is reconstructed, equal to the number of coded masks. The proposed framework is an end-to-end model where the sensing matrix is trained along with the video reconstruction. The encoder learns the binary elements of the sensing matrix and the decoder is trained to recover the unknown video sequence. The reconstruction performance is found to improve when using the trained sensing mask from the network as compared to other mask designs such as random, across a wide variety of compressive sensing reconstruction algorithms. Finally, our analysis and discussion offers insights into understanding the characteristics of the trained mask designs that lead to the improved reconstruction quality. |
| Tasks | Compressive Sensing, Video Compressive Sensing, Video Reconstruction |
| Published | 2016-07-12 |
| URL | http://arxiv.org/abs/1607.03343v2 |
| http://arxiv.org/pdf/1607.03343v2.pdf | |
| PWC | https://paperswithcode.com/paper/deepbinarymask-learning-a-binary-mask-for |
| Repo | |
| Framework | |
Streaming Normalization: Towards Simpler and More Biologically-plausible Normalizations for Online and Recurrent Learning
| Title | Streaming Normalization: Towards Simpler and More Biologically-plausible Normalizations for Online and Recurrent Learning |
| Authors | Qianli Liao, Kenji Kawaguchi, Tomaso Poggio |
| Abstract | We systematically explored a spectrum of normalization algorithms related to Batch Normalization (BN) and propose a generalized formulation that simultaneously solves two major limitations of BN: (1) online learning and (2) recurrent learning. Our proposal is simpler and more biologically-plausible. Unlike previous approaches, our technique can be applied out of the box to all learning scenarios (e.g., online learning, batch learning, fully-connected, convolutional, feedforward, recurrent and mixed — recurrent and convolutional) and compare favorably with existing approaches. We also propose Lp Normalization for normalizing by different orders of statistical moments. In particular, L1 normalization is well-performing, simple to implement, fast to compute, more biologically-plausible and thus ideal for GPU or hardware implementations. |
| Tasks | |
| Published | 2016-10-19 |
| URL | http://arxiv.org/abs/1610.06160v1 |
| http://arxiv.org/pdf/1610.06160v1.pdf | |
| PWC | https://paperswithcode.com/paper/streaming-normalization-towards-simpler-and |
| Repo | |
| Framework | |
