July 28, 2019

2751 words 13 mins read

Paper Group ANR 186

Supervised Learning of Labeled Pointcloud Differences via Cover-Tree Entropy Reduction. The Importance of Automatic Syntactic Features in Vietnamese Named Entity Recognition. Graph Classification via Deep Learning with Virtual Nodes. InverseFaceNet: Deep Monocular Inverse Face Rendering. Adaptive kNN using Expected Accuracy for Classification of Ge …

Supervised Learning of Labeled Pointcloud Differences via Cover-Tree Entropy Reduction


Title	Supervised Learning of Labeled Pointcloud Differences via Cover-Tree Entropy Reduction
Authors	Abraham Smith, Paul Bendich, John Harer, Alex Pieloch, Jay Hineman
Abstract	We introduce a new algorithm, called CDER, for supervised machine learning that merges the multi-scale geometric properties of Cover Trees with the information-theoretic properties of entropy. CDER applies to a training set of labeled pointclouds embedded in a common Euclidean space. If typical pointclouds corresponding to distinct labels tend to differ at any scale in any sub-region, CDER can identify these differences in (typically) linear time, creating a set of distributional coordinates which act as a feature extraction mechanism for supervised learning. We describe theoretical properties and implementation details of CDER, and illustrate its benefits on several synthetic examples.
Tasks
Published	2017-02-26
URL	http://arxiv.org/abs/1702.07959v3
PDF	http://arxiv.org/pdf/1702.07959v3.pdf
PWC	https://paperswithcode.com/paper/supervised-learning-of-labeled-pointcloud
Repo
Framework

The Importance of Automatic Syntactic Features in Vietnamese Named Entity Recognition


Title	The Importance of Automatic Syntactic Features in Vietnamese Named Entity Recognition
Authors	Thai-Hoang Pham, Phuong Le-Hong
Abstract	This paper presents a state-of-the-art system for Vietnamese Named Entity Recognition (NER). By incorporating automatic syntactic features with word embeddings as input for bidirectional Long Short-Term Memory (Bi-LSTM), our system, although simpler than some deep learning architectures, achieves a much better result for Vietnamese NER. The proposed method achieves an overall F1 score of 92.05% on the test set of an evaluation campaign, organized in late 2016 by the Vietnamese Language and Speech Processing (VLSP) community. Our named entity recognition system outperforms the best previous systems for Vietnamese NER by a large margin.
Tasks	Named Entity Recognition, Word Embeddings
Published	2017-05-29
URL	http://arxiv.org/abs/1705.10610v4
PDF	http://arxiv.org/pdf/1705.10610v4.pdf
PWC	https://paperswithcode.com/paper/the-importance-of-automatic-syntactic
Repo
Framework

Graph Classification via Deep Learning with Virtual Nodes


Title	Graph Classification via Deep Learning with Virtual Nodes
Authors	Trang Pham, Truyen Tran, Hoa Dam, Svetha Venkatesh
Abstract	Learning representation for graph classification turns a variable-size graph into a fixed-size vector (or matrix). Such a representation works nicely with algebraic manipulations. Here we introduce a simple method to augment an attributed graph with a virtual node that is bidirectionally connected to all existing nodes. The virtual node represents the latent aspects of the graph, which are not immediately available from the attributes and local connectivity structures. The expanded graph is then put through any node representation method. The representation of the virtual node is then the representation of the entire graph. In this paper, we use the recently introduced Column Network for the expanded graph, resulting in a new end-to-end graph classification model dubbed Virtual Column Network (VCN). The model is validated on two tasks: (i) predicting bio-activity of chemical compounds, and (ii) finding software vulnerability from source code. Results demonstrate that VCN is competitive against well-established rivals.
Tasks	Graph Classification
Published	2017-08-14
URL	http://arxiv.org/abs/1708.04357v1
PDF	http://arxiv.org/pdf/1708.04357v1.pdf
PWC	https://paperswithcode.com/paper/graph-classification-via-deep-learning-with
Repo
Framework

InverseFaceNet: Deep Monocular Inverse Face Rendering


Title	InverseFaceNet: Deep Monocular Inverse Face Rendering
Authors	Hyeongwoo Kim, Michael Zollhöfer, Ayush Tewari, Justus Thies, Christian Richardt, Christian Theobalt
Abstract	We introduce InverseFaceNet, a deep convolutional inverse rendering framework for faces that jointly estimates facial pose, shape, expression, reflectance and illumination from a single input image. By estimating all parameters from just a single image, advanced editing possibilities on a single face image, such as appearance editing and relighting, become feasible in real time. Most previous learning-based face reconstruction approaches do not jointly recover all dimensions, or are severely limited in terms of visual quality. In contrast, we propose to recover high-quality facial pose, shape, expression, reflectance and illumination using a deep neural network that is trained using a large, synthetically created training corpus. Our approach builds on a novel loss function that measures model-space similarity directly in parameter space and significantly improves reconstruction accuracy. We further propose a self-supervised bootstrapping process in the network training loop, which iteratively updates the synthetic training corpus to better reflect the distribution of real-world imagery. We demonstrate that this strategy outperforms completely synthetically trained networks. Finally, we show high-quality reconstructions and compare our approach to several state-of-the-art approaches.
Tasks	Face Reconstruction
Published	2017-03-31
URL	http://arxiv.org/abs/1703.10956v2
PDF	http://arxiv.org/pdf/1703.10956v2.pdf
PWC	https://paperswithcode.com/paper/inversefacenet-deep-monocular-inverse-face
Repo
Framework

Adaptive kNN using Expected Accuracy for Classification of Geo-Spatial Data


Title	Adaptive kNN using Expected Accuracy for Classification of Geo-Spatial Data
Authors	Mark Kibanov, Martin Becker, Juergen Mueller, Martin Atzmueller, Andreas Hotho, Gerd Stumme
Abstract	The k-Nearest Neighbor (kNN) classification approach is conceptually simple - yet widely applied since it often performs well in practical applications. However, using a global constant k does not always provide an optimal solution, e.g., for datasets with an irregular density distribution of data points. This paper proposes an adaptive kNN classifier where k is chosen dynamically for each instance (point) to be classified, such that the expected accuracy of classification is maximized. We define the expected accuracy as the accuracy of a set of structurally similar observations. An arbitrary similarity function can be used to find these observations. We introduce and evaluate different similarity functions. For the evaluation, we use five different classification tasks based on geo-spatial data. Each classification task consists of (tens of) thousands of items. We demonstrate, that the presented expected accuracy measures can be a good estimator for kNN performance, and the proposed adaptive kNN classifier outperforms common kNN and previously introduced adaptive kNN algorithms. Also, we show that the range of considered k can be significantly reduced to speed up the algorithm without negative influence on classification accuracy.
Tasks
Published	2017-12-14
URL	http://arxiv.org/abs/1801.01453v1
PDF	http://arxiv.org/pdf/1801.01453v1.pdf
PWC	https://paperswithcode.com/paper/adaptive-knn-using-expected-accuracy-for
Repo
Framework

A Compression-Inspired Framework for Macro Discovery


Title	A Compression-Inspired Framework for Macro Discovery
Authors	Francisco M. Garcia, Bruno C. da Silva, Philip S. Thomas
Abstract	In this paper we consider the problem of how a reinforcement learning agent tasked with solving a set of related Markov decision processes can use knowledge acquired early in its lifetime to improve its ability to more rapidly solve novel, but related, tasks. One way of exploiting this experience is by identifying recurrent patterns in trajectories obtained from well-performing policies. We propose a three-step framework in which an agent 1) generates a set of candidate open-loop macros by compressing trajectories drawn from near-optimal policies; 2) evaluates the value of each macro; and 3) selects a maximally diverse subset of macros that spans the space of policies typically required for solving the set of related tasks. Our experiments show that extending the original primitive action-set of the agent with the identified macros allows it to more rapidly learn an optimal policy in unseen, but similar MDPs.
Tasks	Efficient Exploration
Published	2017-11-24
URL	http://arxiv.org/abs/1711.09048v3
PDF	http://arxiv.org/pdf/1711.09048v3.pdf
PWC	https://paperswithcode.com/paper/a-compression-inspired-framework-for-macro
Repo
Framework

MoFA: Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction


Title	MoFA: Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction
Authors	Ayush Tewari, Michael Zollhöfer, Hyeongwoo Kim, Pablo Garrido, Florian Bernard, Patrick Pérez, Christian Theobalt
Abstract	In this work we propose a novel model-based deep convolutional autoencoder that addresses the highly challenging problem of reconstructing a 3D human face from a single in-the-wild color image. To this end, we combine a convolutional encoder network with an expert-designed generative model that serves as decoder. The core innovation is our new differentiable parametric decoder that encapsulates image formation analytically based on a generative model. Our decoder takes as input a code vector with exactly defined semantic meaning that encodes detailed face pose, shape, expression, skin reflectance and scene illumination. Due to this new way of combining CNN-based with model-based face reconstruction, the CNN-based encoder learns to extract semantically meaningful parameters from a single monocular input image. For the first time, a CNN encoder and an expert-designed generative model can be trained end-to-end in an unsupervised manner, which renders training on very large (unlabeled) real world data feasible. The obtained reconstructions compare favorably to current state-of-the-art approaches in terms of quality and richness of representation.
Tasks	Face Reconstruction
Published	2017-03-30
URL	http://arxiv.org/abs/1703.10580v2
PDF	http://arxiv.org/pdf/1703.10580v2.pdf
PWC	https://paperswithcode.com/paper/mofa-model-based-deep-convolutional-face
Repo
Framework

Learning Deep Matrix Representations


Title	Learning Deep Matrix Representations
Authors	Kien Do, Truyen Tran, Svetha Venkatesh
Abstract	We present a new distributed representation in deep neural nets wherein the information is represented in native form as a matrix. This differs from current neural architectures that rely on vector representations. We consider matrices as central to the architecture and they compose the input, hidden and output layers. The model representation is more compact and elegant – the number of parameters grows only with the largest dimension of the incoming layer rather than the number of hidden units. We derive several new deep networks: (i) feed-forward nets that map an input matrix into an output matrix, (ii) recurrent nets which map a sequence of input matrices into a sequence of output matrices. We also reinterpret existing models for (iii) memory-augmented networks and (iv) graphs using matrix notations. For graphs we demonstrate how the new notations lead to simple but effective extensions with multiple attentions. Extensive experiments on handwritten digits recognition, face reconstruction, sequence to sequence learning, EEG classification, and graph-based node classification demonstrate the efficacy and compactness of the matrix architectures.
Tasks	EEG, Face Reconstruction, Node Classification
Published	2017-03-04
URL	http://arxiv.org/abs/1703.01454v2
PDF	http://arxiv.org/pdf/1703.01454v2.pdf
PWC	https://paperswithcode.com/paper/learning-deep-matrix-representations
Repo
Framework

On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning


Title	On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning
Authors	Bolin Gao, Lacra Pavel
Abstract	In this paper, we utilize results from convex analysis and monotone operator theory to derive additional properties of the softmax function that have not yet been covered in the existing literature. In particular, we show that the softmax function is the monotone gradient map of the log-sum-exp function. By exploiting this connection, we show that the inverse temperature parameter determines the Lipschitz and co-coercivity properties of the softmax function. We then demonstrate the usefulness of these properties through an application in game-theoretic reinforcement learning.
Tasks
Published	2017-04-03
URL	http://arxiv.org/abs/1704.00805v4
PDF	http://arxiv.org/pdf/1704.00805v4.pdf
PWC	https://paperswithcode.com/paper/on-the-properties-of-the-softmax-function
Repo
Framework

Hardware design for binarization and thinning of fingerprint images


Title	Hardware design for binarization and thinning of fingerprint images
Authors	Farshad Kheiri, Shadrokh Samavi, Nader Karimi
Abstract	Two critical steps in fingerprint recognition are binarization and thinning of the image. The need for real time processing motivates us to select local adaptive thresholding approach for the binarization step. We introduce a new hardware for this purpose based on pipeline architecture. We propose a formula for selecting an optimal block size for the thresholding purpose. To decrease minutiae false detection, the binarized image is dilated. We also present in this paper a new pipeline structure for implementing the thinning algorithm
Tasks
Published	2017-10-12
URL	http://arxiv.org/abs/1710.05749v1
PDF	http://arxiv.org/pdf/1710.05749v1.pdf
PWC	https://paperswithcode.com/paper/hardware-design-for-binarization-and-thinning
Repo
Framework

Video Semantic Object Segmentation by Self-Adaptation of DCNN


Title	Video Semantic Object Segmentation by Self-Adaptation of DCNN
Authors	Seong-Jin Park, Ki-Sang Hong
Abstract	This paper proposes a new framework for semantic segmentation of objects in videos. We address the label inconsistency problem of deep convolutional neural networks (DCNNs) by exploiting the fact that videos have multiple frames; in a few frames the object is confidently-estimated (CE) and we use the information in them to improve labels of the other frames. Given the semantic segmentation results of each frame obtained from DCNN, we sample several CE frames to adapt the DCNN model to the input video by focusing on specific instances in the video rather than general objects in various circumstances. We propose offline and online approaches under different supervision levels. In experiments our method achieved great improvement over the original model and previous state-of-the-art methods.
Tasks	Semantic Segmentation
Published	2017-11-22
URL	http://arxiv.org/abs/1711.08180v1
PDF	http://arxiv.org/pdf/1711.08180v1.pdf
PWC	https://paperswithcode.com/paper/video-semantic-object-segmentation-by-self
Repo
Framework

Using Contexts and Constraints for Improved Geotagging of Human Trafficking Webpages


Title	Using Contexts and Constraints for Improved Geotagging of Human Trafficking Webpages
Authors	Rahul Kapoor, Mayank Kejriwal, Pedro Szekely
Abstract	Extracting geographical tags from webpages is a well-motivated application in many domains. In illicit domains with unusual language models, like human trafficking, extracting geotags with both high precision and recall is a challenging problem. In this paper, we describe a geotag extraction framework in which context, constraints and the openly available Geonames knowledge base work in tandem in an Integer Linear Programming (ILP) model to achieve good performance. In preliminary empirical investigations, the framework improves precision by 28.57% and F-measure by 36.9% on a difficult human trafficking geotagging task compared to a machine learning-based baseline. The method is already being integrated into an existing knowledge base construction system widely used by US law enforcement agencies to combat human trafficking.
Tasks
Published	2017-04-19
URL	http://arxiv.org/abs/1704.05569v1
PDF	http://arxiv.org/pdf/1704.05569v1.pdf
PWC	https://paperswithcode.com/paper/using-contexts-and-constraints-for-improved
Repo
Framework

The (1+$λ$) Evolutionary Algorithm with Self-Adjusting Mutation Rate


Title	The (1+$λ$) Evolutionary Algorithm with Self-Adjusting Mutation Rate
Authors	Benjamin Doerr, Christian Gießen, Carsten Witt, Jing Yang
Abstract	We propose a new way to self-adjust the mutation rate in population-based evolutionary algorithms in discrete search spaces. Roughly speaking, it consists of creating half the offspring with a mutation rate that is twice the current mutation rate and the other half with half the current rate. The mutation rate is then updated to the rate used in that subpopulation which contains the best offspring. We analyze how the $(1+\lambda)$ evolutionary algorithm with this self-adjusting mutation rate optimizes the OneMax test function. We prove that this dynamic version of the $(1+\lambda)$ EA finds the optimum in an expected optimization time (number of fitness evaluations) of $O(n\lambda/\log\lambda+n\log n)$. This time is asymptotically smaller than the optimization time of the classic $(1+\lambda)$ EA. Previous work shows that this performance is best-possible among all $\lambda$-parallel mutation-based unbiased black-box algorithms. This result shows that the new way of adjusting the mutation rate can find optimal dynamic parameter values on the fly. Since our adjustment mechanism is simpler than the ones previously used for adjusting the mutation rate and does not have parameters itself, we are optimistic that it will find other applications.
Tasks
Published	2017-04-07
URL	http://arxiv.org/abs/1704.02191v3
PDF	http://arxiv.org/pdf/1704.02191v3.pdf
PWC	https://paperswithcode.com/paper/the-1-evolutionary-algorithm-with-self
Repo
Framework

DIMAL: Deep Isometric Manifold Learning Using Sparse Geodesic Sampling


Title	DIMAL: Deep Isometric Manifold Learning Using Sparse Geodesic Sampling
Authors	Gautam Pai, Ronen Talmon, Alex Bronstein, Ron Kimmel
Abstract	This paper explores a fully unsupervised deep learning approach for computing distance-preserving maps that generate low-dimensional embeddings for a certain class of manifolds. We use the Siamese configuration to train a neural network to solve the problem of least squares multidimensional scaling for generating maps that approximately preserve geodesic distances. By training with only a few landmarks, we show a significantly improved local and nonlocal generalization of the isometric mapping as compared to analogous non-parametric counterparts. Importantly, the combination of a deep-learning framework with a multidimensional scaling objective enables a numerical analysis of network architectures to aid in understanding their representation power. This provides a geometric perspective to the generalizability of deep learning.
Tasks
Published	2017-11-16
URL	http://arxiv.org/abs/1711.06011v2
PDF	http://arxiv.org/pdf/1711.06011v2.pdf
PWC	https://paperswithcode.com/paper/dimal-deep-isometric-manifold-learning-using
Repo
Framework

AdaBatch: Efficient Gradient Aggregation Rules for Sequential and Parallel Stochastic Gradient Methods


Title	AdaBatch: Efficient Gradient Aggregation Rules for Sequential and Parallel Stochastic Gradient Methods
Authors	Alexandre Défossez, Francis Bach
Abstract	We study a new aggregation operator for gradients coming from a mini-batch for stochastic gradient (SG) methods that allows a significant speed-up in the case of sparse optimization problems. We call this method AdaBatch and it only requires a few lines of code change compared to regular mini-batch SGD algorithms. We provide a theoretical insight to understand how this new class of algorithms is performing and show that it is equivalent to an implicit per-coordinate rescaling of the gradients, similarly to what Adagrad methods can do. In theory and in practice, this new aggregation allows to keep the same sample efficiency of SG methods while increasing the batch size. Experimentally, we also show that in the case of smooth convex optimization, our procedure can even obtain a better loss when increasing the batch size for a fixed number of samples. We then apply this new algorithm to obtain a parallelizable stochastic gradient method that is synchronous but allows speed-up on par with Hogwild! methods as convergence does not deteriorate with the increase of the batch size. The same approach can be used to make mini-batch provably efficient for variance-reduced SG methods such as SVRG.
Tasks
Published	2017-11-06
URL	http://arxiv.org/abs/1711.01761v1
PDF	http://arxiv.org/pdf/1711.01761v1.pdf
PWC	https://paperswithcode.com/paper/adabatch-efficient-gradient-aggregation-rules
Repo
Framework