January 27, 2020

3484 words 17 mins read

Paper Group ANR 1188

Tracking system of Mine Patrol Robot for Low Illumination Environment. Robust Pose Invariant Shape and Texture based Hand Recognition. Dialogue Design and Management for Multi-Session Casual Conversation with Older Adults. Constrained deep neural network architecture search for IoT devices accounting hardware calibration. Recreation of the Periodic …

Tracking system of Mine Patrol Robot for Low Illumination Environment


Title	Tracking system of Mine Patrol Robot for Low Illumination Environment
Authors	Shaoze You, Hua Zhu, Menggang Li, Lei Wang, Chaoquan Tang
Abstract	Computer vision has received a significant attention in recent years, which is one of the important parts for robots to apperceive external environment. Discriminative Correlation Filter (DCF) based trackers gained more popularity due to their efficiency, however, tracking in low-illumination environments is a challenging problem, not yet successfully addressed in the literature. In this work, we tackle the problems by introducing Low-Illumination Long-term Correlation Tracker (LLCT). First, fused features only including HOG and Color Names are employed to boost the tracking efficiency. Second, we used the standard PCA to reduction scheme in the translation and scale estimation phase for accelerating. Third, we learned a long-term correlation filter to keep the long-term memory ability. Finally, update memory templates with interval updates, then re-match existing and initial templates every few frames to maintain template accuracy. The extensive experiments on popular Object Tracking Benchmark OTB-50 datasets have demonstrated that the proposed tracker outperforms the state-of-the-art trackers significantly achieves a high real-time (33FPS) performance. In addition, the proposed approach can be integrated easily in robot system and the running speed performed well. The experimental results show that the novel tracker performance in low-illumination environment is better than that of general trackers.
Tasks	Object Tracking
Published	2019-07-03
URL	https://arxiv.org/abs/1907.01806v3
PDF	https://arxiv.org/pdf/1907.01806v3.pdf
PWC	https://paperswithcode.com/paper/tracking-system-of-mine-patrol-robot-for-low
Repo
Framework

Robust Pose Invariant Shape and Texture based Hand Recognition


Title	Robust Pose Invariant Shape and Texture based Hand Recognition
Authors	F. Sohel, A. El-Sallam, M. Bennamoun
Abstract	This paper presents a novel personal identification and verification system using information extracted from the hand shape and texture. The system has two major constituent modules: a fully automatic and robust peg free segmentation and pose normalisation module, and a recognition module. In the first module, the hand is segmented from its background using a thresholding technique based on Otsu`s method combined with a skin colour detector. A set of fully automatic algorithms are then proposed to segment the palm and fingers. In these algorithms, the skeleton and the contour of the hand and fingers are estimated and used to determine the global pose of the hand and the pose of each individual finger. Finally the palm and fingers are cropped, pose corrected and normalised. In the recognition module, various shape and texture based features are extracted and used for matching purposes. The modified Hausdorff distance, the Iterative Closest Point (ICP) and Independent Component Analysis (ICA) algorithms are used for shape and texture features of the fingers. For the palmprints, we use the Discrete Cosine Transform (DCT), directional line features and ICA. Recognition (identification and verification) tests were performed using fusion strategies based on the similarity scores of the fingers and the palm. Experimental results show that the proposed system exhibits a superior performance over existing systems with an accuracy of over 98% for hand identification and verification (at equal error rate) in a database of 560 different subjects. \|
Tasks
Published	2019-12-22
URL	https://arxiv.org/abs/1912.10373v1
PDF	https://arxiv.org/pdf/1912.10373v1.pdf
PWC	https://paperswithcode.com/paper/robust-pose-invariant-shape-and-texture-based
Repo
Framework

Dialogue Design and Management for Multi-Session Casual Conversation with Older Adults


Title	Dialogue Design and Management for Multi-Session Casual Conversation with Older Adults
Authors	S. Zahra Razavi, Lenhart K. Schubert, Benjamin Kane, Mohammad Rafayet Ali, Kimberly Van Orden, Tianyi Ma
Abstract	We address the problem of designing a conversational avatar capable of a sequence of casual conversations with older adults. Users at risk of loneliness, social anxiety or a sense of ennui may benefit from practicing such conversations in private, at their convenience. We describe an automatic spoken dialogue manager for LISSA, an on-screen virtual agent that can keep older users involved in conversations over several sessions, each lasting 10-20 minutes. The idea behind LISSA is to improve users’ communication skills by providing feedback on their non-verbal behavior at certain points in the course of the conversations. In this paper, we analyze the dialogues collected from the first session between LISSA and each of 8 participants. We examine the quality of the conversations by comparing the transcripts with those collected in a WOZ setting. LISSA’s contributions to the conversations were judged by research assistants who rated the extent to which the contributions were “natural”, “on track”, “encouraging”, “understanding”, “relevant”, and “polite”. The results show that the automatic dialogue manager was able to handle conversation with the users smoothly and naturally.
Tasks
Published	2019-01-20
URL	https://arxiv.org/abs/1901.06620v2
PDF	https://arxiv.org/pdf/1901.06620v2.pdf
PWC	https://paperswithcode.com/paper/dialogue-design-and-management-for-multi
Repo
Framework

Constrained deep neural network architecture search for IoT devices accounting hardware calibration


Title	Constrained deep neural network architecture search for IoT devices accounting hardware calibration
Authors	Florian Scheidegger, Luca Benini, Costas Bekas, Cristiano Malossi
Abstract	Deep neural networks achieve outstanding results in challenging image classification tasks. However, the design of network topologies is a complex task and the research community makes a constant effort in discovering top-accuracy topologies, either manually or employing expensive architecture searches. In this work, we propose a unique narrow-space architecture search that focuses on delivering low-cost and fast executing networks that respect strict memory and time requirements typical of Internet-of-Things (IoT) near-sensor computing platforms. Our approach provides solutions with classification latencies below 10ms running on a $35 device with 1GB RAM and 5.6GFLOPS peak performance. The narrow-space search of floating-point models improves the accuracy on CIFAR10 of an established IoT model from 70.64% to 74.87% respecting the same memory constraints. We further improve the accuracy to 82.07% by including 16-bit half types and we obtain the best accuracy of 83.45% by extending the search with model optimized IEEE 754 reduced types. To the best of our knowledge, we are the first that empirically demonstrate on over 3000 trained models that running with reduced precision pushes the Pareto optimal front by a wide margin. Under a given memory constraint, accuracy is improved by over 7% points for half and over 1% points further for running with the best model individual format.
Tasks	Calibration, Image Classification
Published	2019-09-24
URL	https://arxiv.org/abs/1909.10818v1
PDF	https://arxiv.org/pdf/1909.10818v1.pdf
PWC	https://paperswithcode.com/paper/constrained-deep-neural-network-architecture
Repo
Framework

Recreation of the Periodic Table with an Unsupervised Machine Learning Algorithm


Title	Recreation of the Periodic Table with an Unsupervised Machine Learning Algorithm
Authors	Minoru Kusaba, Chang Liu, Yukinori Koyama, Kiyoyuki Terakura, Ryo Yoshida
Abstract	In 1869, the first draft of the periodic table was published by Russian chemist Dmitri Mendeleev. In terms of data science, his achievement can be viewed as a successful example of feature embedding based on human cognition: chemical properties of all known elements at that time were compressed onto the two-dimensional grid system for tabular display. In this study, we seek to answer the question of whether machine learning can reproduce or recreate the periodic table by using observed physicochemical properties of the elements. To achieve this goal, we developed a periodic table generator (PTG). The PTG is an unsupervised machine learning algorithm based on the generative topographic mapping (GTM), which can automate the translation of high-dimensional data into a tabular form with varying layouts on-demand. The PTG autonomously produced various arrangements of chemical symbols, which organized a two-dimensional array such as Mendeleev’s periodic table or three-dimensional spiral table according to the underlying periodicity in the given data. We further showed what the PTG learned from the element data and how the element features, such as melting point and electronegativity, are compressed to the lower-dimensional latent spaces.
Tasks
Published	2019-12-23
URL	https://arxiv.org/abs/1912.10708v1
PDF	https://arxiv.org/pdf/1912.10708v1.pdf
PWC	https://paperswithcode.com/paper/recreation-of-the-periodic-table-with-an
Repo
Framework

GANFIT: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction


Title	GANFIT: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction
Authors	Baris Gecer, Stylianos Ploumpis, Irene Kotsia, Stefanos Zafeiriou
Abstract	In the past few years, a lot of work has been done towards reconstructing the 3D facial structure from single images by capitalizing on the power of Deep Convolutional Neural Networks (DCNNs). In the most recent works, differentiable renderers were employed in order to learn the relationship between the facial identity features and the parameters of a 3D morphable model for shape and texture. The texture features either correspond to components of a linear texture space or are learned by auto-encoders directly from in-the-wild images. In all cases, the quality of the facial texture reconstruction of the state-of-the-art methods is still not capable of modeling textures in high fidelity. In this paper, we take a radically different approach and harness the power of Generative Adversarial Networks (GANs) and DCNNs in order to reconstruct the facial texture and shape from single images. That is, we utilize GANs to train a very powerful generator of facial texture in UV space. Then, we revisit the original 3D Morphable Models (3DMMs) fitting approaches making use of non-linear optimization to find the optimal latent parameters that best reconstruct the test image but under a new perspective. We optimize the parameters with the supervision of pretrained deep identity features through our end-to-end differentiable framework. We demonstrate excellent results in photorealistic and identity preserving 3D face reconstructions and achieve for the first time, to the best of our knowledge, facial texture reconstruction with high-frequency details.
Tasks	3D Face Reconstruction, Face Reconstruction
Published	2019-02-15
URL	http://arxiv.org/abs/1902.05978v2
PDF	http://arxiv.org/pdf/1902.05978v2.pdf
PWC	https://paperswithcode.com/paper/ganfit-generative-adversarial-network-fitting
Repo
Framework

Learning Hierarchical Control for Robust In-Hand Manipulation


Title	Learning Hierarchical Control for Robust In-Hand Manipulation
Authors	Tingguang Li, Krishnan Srinivasan, Max Qing-Hu Meng, Wenzhen Yuan, Jeannette Bohg
Abstract	Robotic in-hand manipulation has been a long-standing challenge due to the complexity of modelling hand and object in contact and of coordinating finger motion for complex manipulation sequences. To address these challenges, the majority of prior work has either focused on model-based, low-level controllers or on model-free deep reinforcement learning that each have their own limitations. We propose a hierarchical method that relies on traditional, model-based controllers on the low-level and learned policies on the mid-level. The low-level controllers can robustly execute different manipulation primitives (reposing, sliding, flipping). The mid-level policy orchestrates these primitives. We extensively evaluate our approach in simulation with a 3-fingered hand that controls three degrees of freedom of elongated objects. We show that our approach can move objects between almost all the possible poses in the workspace while keeping them firmly grasped. We also show that our approach is robust to inaccuracies in the object models and to observation noise. Finally, we show how our approach generalizes to objects of other shapes.
Tasks
Published	2019-10-24
URL	https://arxiv.org/abs/1910.10985v1
PDF	https://arxiv.org/pdf/1910.10985v1.pdf
PWC	https://paperswithcode.com/paper/learning-hierarchical-control-for-robust-in
Repo
Framework

Low-rank matrix completion and denoising under Poisson noise


Title	Low-rank matrix completion and denoising under Poisson noise
Authors	Andrew D. McRae, Mark A. Davenport
Abstract	This paper considers the problem of estimating a low-rank matrix from the observation of all, or a subset, of its entries in the presence of Poisson noise. When we observe all the entries, this is a problem of matrix denoising; when we observe only a subset of the entries, this is a problem of matrix completion. In both cases, we exploit an assumption that the underlying matrix is low-rank. Specifically, we analyze several estimators, including a constrained nuclear-norm minimization program, nuclear-norm regularized least squares, and a nonconvex constrained low-rank optimization problem. We show that for all three estimators, with high probability, we have an upper error bound (in the Frobenius norm error metric) that depends on the matrix rank, the fraction of the elements observed, and maximal row and column sums of the true matrix. We furthermore show that the above results are minimax optimal (within a universal constant) in classes of matrices with low rank and bounded row and column sums. We also extend these results to handle the case of matrix multinomial denoising and completion.
Tasks	Denoising, Low-Rank Matrix Completion, Matrix Completion
Published	2019-07-11
URL	https://arxiv.org/abs/1907.05325v1
PDF	https://arxiv.org/pdf/1907.05325v1.pdf
PWC	https://paperswithcode.com/paper/low-rank-matrix-completion-and-denoising
Repo
Framework

A divide-and-conquer algorithm for binary matrix completion


Title	A divide-and-conquer algorithm for binary matrix completion
Authors	Melanie Beckerleg, Andrew Thompson
Abstract	We propose an algorithm for low rank matrix completion for matrices with binary entries which obtains explicit binary factors. Our algorithm, which we call TBMC (\emph{Tiling for Binary Matrix Completion}), gives interpretable output in the form of binary factors which represent a decomposition of the matrix into tiles. Our approach is inspired by a popular algorithm from the data mining community called PROXIMUS: it adopts the same recursive partitioning approach while extending to missing data. The algorithm relies upon rank-one approximations of incomplete binary matrices, and we propose a linear programming (LP) approach for solving this subproblem. We also prove a $2$-approximation result for the LP approach which holds for any level of subsampling and for any subsampling pattern. Our numerical experiments show that TBMC outperforms existing methods on recommender systems arising in the context of real datasets.
Tasks	Low-Rank Matrix Completion, Matrix Completion, Recommendation Systems
Published	2019-07-09
URL	https://arxiv.org/abs/1907.04251v1
PDF	https://arxiv.org/pdf/1907.04251v1.pdf
PWC	https://paperswithcode.com/paper/a-divide-and-conquer-algorithm-for-binary
Repo
Framework

Transfer Learning for Sequence Labeling Using Source Model and Target Data


Title	Transfer Learning for Sequence Labeling Using Source Model and Target Data
Authors	Lingzhen Chen, Alessandro Moschitti
Abstract	In this paper, we propose an approach for transferring the knowledge of a neural model for sequence labeling, learned from the source domain, to a new model trained on a target domain, where new label categories appear. Our transfer learning (TL) techniques enable to adapt the source model using the target data and new categories, without accessing to the source data. Our solution consists in adding new neurons in the output layer of the target model and transferring parameters from the source model, which are then fine-tuned with the target data. Additionally, we propose a neural adapter to learn the difference between the source and the target label distribution, which provides additional important information to the target model. Our experiments on Named Entity Recognition show that (i) the learned knowledge in the source model can be effectively transferred when the target data contains new categories and (ii) our neural adapter further improves such transfer.
Tasks	Named Entity Recognition, Transfer Learning
Published	2019-02-14
URL	http://arxiv.org/abs/1902.05309v1
PDF	http://arxiv.org/pdf/1902.05309v1.pdf
PWC	https://paperswithcode.com/paper/transfer-learning-for-sequence-labeling-using
Repo
Framework

Weight Map Layer for Noise and Adversarial Attack Robustness


Title	Weight Map Layer for Noise and Adversarial Attack Robustness
Authors	Mohammed Amer, Tomás Maul
Abstract	Convolutional neural networks (CNNs) are known for their good performance and generalization in vision-related tasks and have become state-of-the-art in both application and research-based domains. However, just like other neural network models, they suffer from a susceptibility to noise and adversarial attacks. An adversarial defence aims at reducing a neural network’s susceptibility to adversarial attacks through learning or architectural modifications. We propose a weight map layer (WM) as a generic architectural addition to CNNs and show that it can increase their robustness to noise and adversarial attacks. We further explain the enhanced robustness of the two WM variants introduced via an adaptive noise-variance amplification (ANVA) hypothesis and provide evidence and insights in support of it. We show that the WM layer can be integrated into scaled up models to increase their noise and adversarial attack robustness, while achieving the same or similar accuracy levels.
Tasks	Adversarial Attack
Published	2019-05-02
URL	http://arxiv.org/abs/1905.00568v1
PDF	http://arxiv.org/pdf/1905.00568v1.pdf
PWC	https://paperswithcode.com/paper/weight-map-layer-for-noise-and-adversarial
Repo
Framework

Depth Restoration: A fast low-rank matrix completion via dual-graph regularization


Title	Depth Restoration: A fast low-rank matrix completion via dual-graph regularization
Authors	Wenxiang Zuo, Qiang Li, Xianming Liu
Abstract	As a real scenes sensing approach, depth information obtains the widespread applications. However, resulting from the restriction of depth sensing technology, the depth map captured in practice usually suffers terrible noise and missing values at plenty of pixels. In this paper, we propose a fast low-rank matrix completion via dual-graph regularization for depth restoration. Specifically, the depth restoration can be transformed into a low-rank matrix completion problem. In order to complete the low-rank matrix and restore it to the depth map, the proposed dual-graph method containing the local and non-local graph regularizations exploits the local similarity of depth maps and the gradient consistency of depth-color counterparts respectively. In addition, the proposed approach achieves the high speed depth restoration due to closed-form solution. Experimental results demonstrate that the proposed method outperforms the state-of-the-art methods with respect to both objective and subjective quality evaluations, especially for serious depth degeneration.
Tasks	Low-Rank Matrix Completion, Matrix Completion
Published	2019-07-05
URL	https://arxiv.org/abs/1907.02841v4
PDF	https://arxiv.org/pdf/1907.02841v4.pdf
PWC	https://paperswithcode.com/paper/depth-restoration-a-fast-low-rank-matrix
Repo
Framework

Efficiently escaping saddle points on manifolds


Title	Efficiently escaping saddle points on manifolds
Authors	Chris Criscitiello, Nicolas Boumal
Abstract	Smooth, non-convex optimization problems on Riemannian manifolds occur in machine learning as a result of orthonormality, rank or positivity constraints. First- and second-order necessary optimality conditions state that the Riemannian gradient must be zero, and the Riemannian Hessian must be positive semidefinite. Generalizing Jin et al.‘s recent work on perturbed gradient descent (PGD) for optimization on linear spaces [How to Escape Saddle Points Efficiently (2017), Stochastic Gradient Descent Escapes Saddle Points Efficiently (2019)], we propose a version of perturbed Riemannian gradient descent (PRGD) to show that necessary optimality conditions can be met approximately with high probability, without evaluating the Hessian. Specifically, for an arbitrary Riemannian manifold $\mathcal{M}$ of dimension $d$, a sufficiently smooth (possibly non-convex) objective function $f$, and under weak conditions on the retraction chosen to move on the manifold, with high probability, our version of PRGD produces a point with gradient smaller than $\epsilon$ and Hessian within $\sqrt{\epsilon}$ of being positive semidefinite in $O((\log{d})^4 / \epsilon^{2})$ gradient queries. This matches the complexity of PGD in the Euclidean case. Crucially, the dependence on dimension is low. This matters for large-scale applications including PCA and low-rank matrix completion, which both admit natural formulations on manifolds. The key technical idea is to generalize PRGD with a distinction between two types of gradient steps: “steps on the manifold” and “perturbed steps in a tangent space of the manifold.” Ultimately, this distinction makes it possible to extend Jin et al.‘s analysis seamlessly.
Tasks	Low-Rank Matrix Completion, Matrix Completion
Published	2019-06-10
URL	https://arxiv.org/abs/1906.04321v3
PDF	https://arxiv.org/pdf/1906.04321v3.pdf
PWC	https://paperswithcode.com/paper/efficiently-escaping-saddle-points-on
Repo
Framework

TopicSifter: Interactive Search Space Reduction Through Targeted Topic Modeling


Title	TopicSifter: Interactive Search Space Reduction Through Targeted Topic Modeling
Authors	Hannah Kim, Dongjin Choi, Barry Drake, Alex Endert, Haesun Park
Abstract	Topic modeling is commonly used to analyze and understand large document collections. However, in practice, users want to focus on specific aspects or “targets” rather than the entire corpus. For example, given a large collection of documents, users may want only a smaller subset which more closely aligns with their interests, tasks, and domains. In particular, our paper focuses on large-scale document retrieval with high recall where any missed relevant documents can be critical. A simple keyword matching search is generally not effective nor efficient as 1) it is difficult to find a list of keyword queries that can cover the documents of interest before exploring the dataset, 2) some documents may not contain the exact keywords of interest but may still be highly relevant, and 3) some words have multiple meanings, which would result in irrelevant documents included in the retrieved subset. In this paper, we present TopicSifter, a visual analytics system for interactive search space reduction. Our system utilizes targeted topic modeling based on nonnegative matrix factorization and allows users to give relevance feedback in order to refine their target and guide the topic modeling to the most relevant results.
Tasks
Published	2019-07-28
URL	https://arxiv.org/abs/1907.12079v1
PDF	https://arxiv.org/pdf/1907.12079v1.pdf
PWC	https://paperswithcode.com/paper/topicsifter-interactive-search-space
Repo
Framework

On importance-weighted autoencoders


Title	On importance-weighted autoencoders
Authors	Axel Finke, Alexandre H. Thiery
Abstract	The importance weighted autoencoder (IWAE) (Burda et al., 2016) is a popular variational-inference method which achieves a tighter evidence bound (and hence a lower bias) than standard variational autoencoders by optimising a multi-sample objective, i.e. an objective that is expressible as an integral over $K > 1$ Monte Carlo samples. Unfortunately, IWAE crucially relies on the availability of reparametrisations and even if these exist, the multi-sample objective leads to inference-network gradients which break down as $K$ is increased (Rainforth et al., 2018). This breakdown can only be circumvented by removing high-variance score-function terms, either by heuristically ignoring them (which yields the ‘sticking-the-landing’ IWAE (IWAE-STL) gradient from Roeder et al. (2017)) or through an identity from Tucker et al. (2019) (which yields the ‘doubly-reparametrised’ IWAE (IWAE-DREG) gradient). In this work, we argue that directly optimising the proposal distribution in importance sampling as in the reweighted wake-sleep (RWS) algorithm from Bornschein & Bengio (2015) is preferable to optimising IWAE-type multi-sample objectives. To formalise this argument, we introduce an adaptive-importance sampling framework termed adaptive importance sampling for learning (AISLE) which slightly generalises the RWS algorithm. We then show that AISLE admits IWAE-STL and IWAE-DREG (i.e. the IWAE-gradients which avoid breakdown) as special cases.
Tasks
Published	2019-07-24
URL	https://arxiv.org/abs/1907.10477v2
PDF	https://arxiv.org/pdf/1907.10477v2.pdf
PWC	https://paperswithcode.com/paper/on-the-relationship-between-variational
Repo
Framework