January 26, 2020

3263 words 16 mins read

Paper Group ANR 1488

Iteratively reweighted penalty alternating minimization methods with continuation for image deblurring. DeepView: View Synthesis with Learned Gradient Descent. Improving Interactive Reinforcement Agent Planning with Human Demonstration. Multi-Cue Vehicle Detection for Semantic Video Compression In Georegistered Aerial Videos. Knowledge transfer in …

Iteratively reweighted penalty alternating minimization methods with continuation for image deblurring


Title	Iteratively reweighted penalty alternating minimization methods with continuation for image deblurring
Authors	Tao Sun, Dongsheng Li, Hao Jiang, Zhe Quan
Abstract	In this paper, we consider a class of nonconvex problems with linear constraints appearing frequently in the area of image processing. We solve this problem by the penalty method and propose the iteratively reweighted alternating minimization algorithm. To speed up the algorithm, we also apply the continuation strategy to the penalty parameter. A convergence result is proved for the algorithm. Compared with the nonconvex ADMM, the proposed algorithm enjoys both theoretical and computational advantages like weaker convergence requirements and faster speed. Numerical results demonstrate the efficiency of the proposed algorithm.
Tasks	Deblurring
Published	2019-02-09
URL	http://arxiv.org/abs/1902.04062v1
PDF	http://arxiv.org/pdf/1902.04062v1.pdf
PWC	https://paperswithcode.com/paper/iteratively-reweighted-penalty-alternating
Repo
Framework

DeepView: View Synthesis with Learned Gradient Descent


Title	DeepView: View Synthesis with Learned Gradient Descent
Authors	John Flynn, Michael Broxton, Paul Debevec, Matthew DuVall, Graham Fyffe, Ryan Overbeck, Noah Snavely, Richard Tucker
Abstract	We present a novel approach to view synthesis using multiplane images (MPIs). Building on recent advances in learned gradient descent, our algorithm generates an MPI from a set of sparse camera viewpoints. The resulting method incorporates occlusion reasoning, improving performance on challenging scene features such as object boundaries, lighting reflections, thin structures, and scenes with high depth complexity. We show that our method achieves high-quality, state-of-the-art results on two datasets: the Kalantari light field dataset, and a new camera array dataset, Spaces, which we make publicly available.
Tasks
Published	2019-06-18
URL	https://arxiv.org/abs/1906.07316v1
PDF	https://arxiv.org/pdf/1906.07316v1.pdf
PWC	https://paperswithcode.com/paper/deepview-view-synthesis-with-learned-gradient-1
Repo
Framework

Improving Interactive Reinforcement Agent Planning with Human Demonstration


Title	Improving Interactive Reinforcement Agent Planning with Human Demonstration
Authors	Guangliang Li, Randy Gomez, Keisuke Nakamura, Jinying Lin, Qilei Zhang, Bo He
Abstract	TAMER has proven to be a powerful interactive reinforcement learning method for allowing ordinary people to teach and personalize autonomous agents’ behavior by providing evaluative feedback. However, a TAMER agent planning with UCT—a Monte Carlo Tree Search strategy, can only update states along its path and might induce high learning cost especially for a physical robot. In this paper, we propose to drive the agent’s exploration along the optimal path and reduce the learning cost by initializing the agent’s reward function via inverse reinforcement learning from demonstration. We test our proposed method in the RL benchmark domain—Grid World—with different discounts on human reward. Our results show that learning from demonstration can allow a TAMER agent to learn a roughly optimal policy up to the deepest search and encourage the agent to explore along the optimal path. In addition, we find that learning from demonstration can improve the learning efficiency by reducing total feedback, the number of incorrect actions and increasing the ratio of correct actions to obtain an optimal policy, allowing a TAMER agent to converge faster.
Tasks
Published	2019-04-18
URL	http://arxiv.org/abs/1904.08621v1
PDF	http://arxiv.org/pdf/1904.08621v1.pdf
PWC	https://paperswithcode.com/paper/improving-interactive-reinforcement-agent
Repo
Framework

Multi-Cue Vehicle Detection for Semantic Video Compression In Georegistered Aerial Videos


Title	Multi-Cue Vehicle Detection for Semantic Video Compression In Georegistered Aerial Videos
Authors	Noor Al-Shakarji, Filiz Bunyak, Hadi Aliakbarpour, Guna Seetharaman, Kannappan Palaniappan
Abstract	Detection of moving objects such as vehicles in videos acquired from an airborne camera is very useful for video analytics applications. Using fast low power algorithms for onboard moving object detection would also provide region of interest-based semantic information for scene content aware image compression. This would enable more efficient and flexible communication link utilization in lowbandwidth airborne cloud computing networks. Despite recent advances in both UAV or drone platforms and imaging sensor technologies, vehicle detection from aerial video remains challenging due to small object sizes, platform motion and camera jitter, obscurations, scene complexity and degraded imaging conditions. This paper proposes an efficient moving vehicle detection pipeline which synergistically fuses both appearance and motion-based detections in a complementary manner using deep learning combined with flux tensor spatio-temporal filtering. Our proposed multi-cue pipeline is able to detect moving vehicles with high precision and recall, while filtering out false positives such as parked vehicles, through intelligent fusion. Experimental results show that incorporating contextual information of moving vehicles enables high semantic compression ratios of over 100:1 with high image fidelity, for better utilization of limited bandwidth air-to-ground network links.
Tasks	Image Compression, Object Detection, Video Compression
Published	2019-07-02
URL	https://arxiv.org/abs/1907.01176v1
PDF	https://arxiv.org/pdf/1907.01176v1.pdf
PWC	https://paperswithcode.com/paper/multi-cue-vehicle-detection-for-semantic
Repo
Framework

Knowledge transfer in deep block-modular neural networks


Title	Knowledge transfer in deep block-modular neural networks
Authors	Alexander V. Terekhov, Guglielmo Montone, J. Kevin O’Regan
Abstract	Although deep neural networks (DNNs) have demonstrated impressive results during the last decade, they remain highly specialized tools, which are trained – often from scratch – to solve each particular task. The human brain, in contrast, significantly re-uses existing capacities when learning to solve new tasks. In the current study we explore a block-modular architecture for DNNs, which allows parts of the existing network to be re-used to solve a new task without a decrease in performance when solving the original task. We show that networks with such architectures can outperform networks trained from scratch, or perform comparably, while having to learn nearly 10 times fewer weights than the networks trained from scratch.
Tasks	Transfer Learning
Published	2019-07-24
URL	https://arxiv.org/abs/1908.08017v1
PDF	https://arxiv.org/pdf/1908.08017v1.pdf
PWC	https://paperswithcode.com/paper/knowledge-transfer-in-deep-block-modular
Repo
Framework

Label-similarity Curriculum Learning


Title	Label-similarity Curriculum Learning
Authors	Urun Dogan, Aniket Anand Deshmukh, Marcin Machura, Christian Igel
Abstract	Curriculum learning can improve neural network training by guiding the optimization to desirable optima. We propose a novel curriculum learning approach for image classification that adapts the loss function by changing the label representation. The idea is to use a probability distribution over classes as target label, where the class probabilities reflect the similarity to the true class. Gradually, this label representation is shifted towards the standard one-hot-encoding. That is, in the beginning minor mistakes are corrected less than large mistakes, resembling a teaching process in which broad concepts are explained first before subtle differences are taught. The class similarity can be based on prior knowledge. For the special case of the labels being natural words, we propose a generic way to automatically compute the similarities. The natural words are embedded into Euclidean space using a standard word embedding. The probability of each class is then a function of the cosine similarity between the vector representations of the class and the true label. The proposed label-similarity curriculum learning (LCL) approach was empirically evaluated on several popular deep learning architectures for image classification task applied to three datasets, ImageNet, CIFAR100, and AWA2. In all scenarios, LCL was able to improve the classification accuracy on the test data compared to standard training.
Tasks	Image Classification
Published	2019-11-15
URL	https://arxiv.org/abs/1911.06902v1
PDF	https://arxiv.org/pdf/1911.06902v1.pdf
PWC	https://paperswithcode.com/paper/label-similarity-curriculum-learning
Repo
Framework

Fingerprint Synthesis: Search with 100 Million Prints


Title	Fingerprint Synthesis: Search with 100 Million Prints
Authors	Vishesh Mistry, Joshua J. Engelsma, Anil K. Jain
Abstract	Evaluation of large-scale fingerprint search algorithms has been limited due to lack of publicly available datasets. A solution to this problem is to synthesize a dataset of fingerprints with characteristics similar to those of real fingerprints. We propose a Generative Adversarial Network (GAN) to synthesize a fingerprint dataset consisting of 100 million fingerprint images. In comparison to published methods, our approach incorporates an identity loss which guides the generator to synthesize a diverse set of fingerprints corresponding to more distinct identities. To demonstrate that the characteristics of our synthesized fingerprints are similar to those of real fingerprints, we show that (i) the NFIQ quality value distribution of the synthetic fingerprints follows the corresponding distribution of real fingerprints and (ii) the synthetic fingerprints are more distinct than existing synthetic fingerprints (and more closely align with the distinctiveness of real fingerprints). We use our synthesis algorithm to generate 100 million fingerprint images in 17.5 hours on 100 Tesla K80 GPUs when executed in parallel. Finally, we report for the first time in open literature, search accuracy (DeepPrint rank-1 accuracy of 91.4%) against a gallery of 100 million fingerprint images (using 2,000 NIST SD4 rolled prints as the queries).
Tasks
Published	2019-12-16
URL	https://arxiv.org/abs/1912.07195v1
PDF	https://arxiv.org/pdf/1912.07195v1.pdf
PWC	https://paperswithcode.com/paper/fingerprint-synthesis-search-with-100-million
Repo
Framework

Achieving the Bayes Error Rate in Synchronization and Block Models by SDP, Robustly


Title	Achieving the Bayes Error Rate in Synchronization and Block Models by SDP, Robustly
Authors	Yingjie Fei, Yudong Chen
Abstract	We study the statistical performance of semidefinite programming (SDP) relaxations for clustering under random graph models. Under the $\mathbb{Z}_{2}$ Synchronization model, Censored Block Model and Stochastic Block Model, we show that SDP achieves an error rate of the form [ \exp\Big[-\big(1-o(1)\big)\bar{n} I^* \Big]. ] Here $\bar{n}$ is an appropriate multiple of the number of nodes and $I^*$ is an information-theoretic measure of the signal-to-noise ratio. We provide matching lower bounds on the Bayes error for each model and therefore demonstrate that the SDP approach is Bayes optimal. As a corollary, our results imply that SDP achieves the optimal exact recovery threshold under each model. Furthermore, we show that SDP is robust: the above bound remains valid under semirandom versions of the models in which the observed graph is modified by a monotone adversary. Our proof is based on a novel primal-dual analysis of SDP under a unified framework for all three models, and the analysis shows that SDP tightly approximates a joint majority voting procedure.
Tasks
Published	2019-04-21
URL	http://arxiv.org/abs/1904.09635v1
PDF	http://arxiv.org/pdf/1904.09635v1.pdf
PWC	https://paperswithcode.com/paper/achieving-the-bayes-error-rate-in
Repo
Framework

Research Frontiers in Transfer Learning – a systematic and bibliometric review


Title	Research Frontiers in Transfer Learning – a systematic and bibliometric review
Authors	Frederico Guth, Teofilo Emidio de-Campos
Abstract	Humans can learn from very few samples, demonstrating an outstanding generalization ability that learning algorithms are still far from reaching. Currently, the most successful models demand enormous amounts of well-labeled data, which are expensive and difficult to obtain, becoming one of the biggest obstacles to the use of machine learning in practice. This scenario shows the massive potential for Transfer Learning, which aims to harness previously acquired knowledge to the learning of new tasks more effectively and efficiently. In this systematic review, we apply a quantitative method to select the main contributions to the field and make use of bibliographic coupling metrics to identify research frontiers. We further analyze the linguistic variation between the classics of the field and the frontier and map promising research directions.
Tasks	Transfer Learning
Published	2019-12-18
URL	https://arxiv.org/abs/1912.08812v1
PDF	https://arxiv.org/pdf/1912.08812v1.pdf
PWC	https://paperswithcode.com/paper/research-frontiers-in-transfer-learning-a
Repo
Framework

Precise Proximal Femur Fracture Classification for Interactive Training and Surgical Planning


Title	Precise Proximal Femur Fracture Classification for Interactive Training and Surgical Planning
Authors	Amelia Jiménez-Sánchez, Anees Kazi, Shadi Albarqouni, Chlodwig Kirchhoff, Peter Biberthaler, Nassir Navab, Sonja Kirchhoff, Diana Mateus
Abstract	We demonstrate the feasibility of a fully automatic computer-aided diagnosis (CAD) tool, based on deep learning, that localizes and classifies proximal femur fractures on X-ray images according to the AO classification. The proposed framework aims to improve patient treatment planning and provide support for the training of trauma surgeon residents. A database of 1347 clinical radiographic studies was collected. Radiologists and trauma surgeons annotated all fractures with bounding boxes, and provided a classification according to the AO standard. The proposed CAD tool for the classification of radiographs into types “A”, “B” and “not-fractured”, reaches a F1-score of 87% and AUC of 0.95, when classifying fractures versus not-fractured cases it improves up to 94% and 0.98. Prior localization of the fracture results in an improvement with respect to full image classification. 100% of the predicted centers of the region of interest are contained in the manually provided bounding boxes. The system retrieves on average 9 relevant images (from the same class) out of 10 cases. Our CAD scheme localizes, detects and further classifies proximal femur fractures achieving results comparable to expert-level and state-of-the-art performance. Our auxiliary localization model was highly accurate predicting the region of interest in the radiograph. We further investigated several strategies of verification for its adoption into the daily clinical routine. A sensitivity analysis of the size of the ROI and image retrieval as a clinical use case were presented.
Tasks	Image Classification, Image Retrieval
Published	2019-02-04
URL	https://arxiv.org/abs/1902.01338v2
PDF	https://arxiv.org/pdf/1902.01338v2.pdf
PWC	https://paperswithcode.com/paper/towards-an-interactive-and-interpretable-cad
Repo
Framework

A Convergence Result for Regularized Actor-Critic Methods


Title	A Convergence Result for Regularized Actor-Critic Methods
Authors	Wesley Suttle, Zhuoran Yang, Kaiqing Zhang, Ji Liu
Abstract	In this paper, we present a probability one convergence proof, under suitable conditions, of a certain class of actor-critic algorithms for finding approximate solutions to entropy-regularized MDPs using the machinery of stochastic approximation. To obtain this overall result, we prove the convergence of policy evaluation with general regularizers when using linear approximation architectures and show convergence of entropy-regularized policy improvement.
Tasks
Published	2019-07-13
URL	https://arxiv.org/abs/1907.06138v2
PDF	https://arxiv.org/pdf/1907.06138v2.pdf
PWC	https://paperswithcode.com/paper/stochastic-convergence-results-for
Repo
Framework

Compressive-Sensing Data Reconstruction for Structural Health Monitoring: A Machine-Learning Approach


Title	Compressive-Sensing Data Reconstruction for Structural Health Monitoring: A Machine-Learning Approach
Authors	Yuequan Bao, Zhiyi Tang, Hui Li
Abstract	Compressive sensing (CS) has been studied and applied in structural health monitoring for wireless data acquisition and transmission, structural modal identification, and spare damage identification. The key issue in CS is finding the optimal solution for sparse optimization. In the past years, many algorithms have been proposed in the field of applied mathematics. In this paper, we propose a machine-learning-based approach to solve the CS data-reconstruction problem. By treating a computation process as a data flow, the process of CS-based data reconstruction is formalized into a standard supervised-learning task. The prior knowledge, i.e., the basis matrix and the CS-sampled signals, are used as the input and the target of the network; the basis coefficient matrix is embedded as the parameters of a certain layer; the objective function of conventional compressive sensing is set as the loss function of the network. Regularized by l1-norm, these basis coefficients are optimized to reduce the error between the original CS-sampled signals and the masked reconstructed signals with a common optimization algorithm. Also, the proposed network can handle complex bases, such as a Fourier basis. Benefiting from the nature of a multi-neuron layer, multiple signal channels can be reconstructed simultaneously. Meanwhile, the disassembled use of a large-scale basis makes the method memory-efficient. A numerical example of multiple sinusoidal waves and an example of field-test wireless data from a suspension bridge are carried out to illustrate the data-reconstruction ability of the proposed approach. The results show that high reconstruction accuracy can be obtained by the machine learning-based approach. Also, the parameters of the network have clear meanings; the inference of the mapping between input and output is fully transparent, making the CS data reconstruction neural network interpretable.
Tasks	Compressive Sensing
Published	2019-01-07
URL	http://arxiv.org/abs/1901.01995v2
PDF	http://arxiv.org/pdf/1901.01995v2.pdf
PWC	https://paperswithcode.com/paper/compressive-sensing-data-reconstruction-for
Repo
Framework

Adverse Childhood Experiences Ontology for Mental Health Surveillance, Research, and Evaluation: Advanced Knowledge Representation and Semantic Web Techniques


Title	Adverse Childhood Experiences Ontology for Mental Health Surveillance, Research, and Evaluation: Advanced Knowledge Representation and Semantic Web Techniques
Authors	Jon Hael Brenas, Eun Kyong Shin, Arash Shaban-Nejad
Abstract	Background: Adverse Childhood Experiences (ACEs), a set of negative events and processes that a person might encounter during childhood and adolescence, have been proven to be linked to increased risks of a multitude of negative health outcomes and conditions when children reach adulthood and beyond. Objective: To better understand the relationship between ACEs and their relevant risk factors with associated health outcomes and to eventually design and implement preventive interventions, access to an integrated coherent dataset is needed. Therefore, we implemented a formal ontology as a resource to allow the mental health community to facilitate data integration and knowledge modeling and to improve ACEs surveillance and research. Methods: We use advanced knowledge representation and Semantic Web tools and techniques to implement the ontology. The current implementation of the ontology is expressed in the description logic ALCRIQ(D), a sublogic of Web Ontology Language (OWL 2). Results: The ACEs Ontology has been implemented and made available to the mental health community and the public via the BioPortal repository. Moreover, multiple use-case scenarios have been introduced to showcase and evaluate the usability of the ontology in action. The ontology was created to be used by major actors in the ACEs community with different applications, from the diagnosis of individuals and predicting potential negative outcomes that they might encounter to the prevention of ACEs in a population and designing interventions and policies. Conclusions: The ACEs Ontology provides a uniform and reusable semantic network and an integrated knowledge structure for mental health practitioners and researchers to improve ACEs surveillance and evaluation.
Tasks
Published	2019-11-19
URL	https://arxiv.org/abs/1912.05530v1
PDF	https://arxiv.org/pdf/1912.05530v1.pdf
PWC	https://paperswithcode.com/paper/adverse-childhood-experiences-ontology-for
Repo
Framework

Knowledge Graph Fact Prediction via Knowledge-Enriched Tensor Factorization


Title	Knowledge Graph Fact Prediction via Knowledge-Enriched Tensor Factorization
Authors	Ankur Padia, Kostantinos Kalpakis, Francis Ferraro, Tim Finin
Abstract	We present a family of novel methods for embedding knowledge graphs into real-valued tensors. These tensor-based embeddings capture the ordered relations that are typical in the knowledge graphs represented by semantic web languages like RDF. Unlike many previous models, our methods can easily use prior background knowledge provided by users or extracted automatically from existing knowledge graphs. In addition to providing more robust methods for knowledge graph embedding, we provide a provably-convergent, linear tensor factorization algorithm. We demonstrate the efficacy of our models for the task of predicting new facts across eight different knowledge graphs, achieving between 5% and 50% relative improvement over existing state-of-the-art knowledge graph embedding techniques. Our empirical evaluation shows that all of the tensor decomposition models perform well when the average degree of an entity in a graph is high, with constraint-based models doing better on graphs with a small number of highly similar relations and regularization-based models dominating for graphs with relations of varying degrees of similarity.
Tasks	Graph Embedding, Knowledge Graph Embedding, Knowledge Graphs
Published	2019-02-08
URL	http://arxiv.org/abs/1902.03077v1
PDF	http://arxiv.org/pdf/1902.03077v1.pdf
PWC	https://paperswithcode.com/paper/knowledge-graph-fact-prediction-via-knowledge
Repo
Framework

SentiMATE: Learning to play Chess through Natural Language Processing


Title	SentiMATE: Learning to play Chess through Natural Language Processing
Authors	Isaac Kamlish, Isaac Bentata Chocron, Nicholas McCarthy
Abstract	We present SentiMATE, a novel end-to-end Deep Learning model for Chess, employing Natural Language Processing that aims to learn an effective evaluation function assessing move quality. This function is pre-trained on the sentiment of commentary associated with the training moves and is used to guide and optimize the agent’s game-playing decision making. The contributions of this research are three-fold: we build and put forward both a classifier which extracts commentary describing the quality of Chess moves in vast commentary datasets, and a Sentiment Analysis model trained on Chess commentary to accurately predict the quality of said moves, to then use those predictions to evaluate the optimal next move of a Chess agent. Both classifiers achieve over 90 % classification accuracy. Lastly, we present a Chess engine, SentiMATE, which evaluates Chess moves based on a pre-trained sentiment evaluation function. Our results exhibit strong evidence to support our initial hypothesis - “Can Natural Language Processing be used to train a novel and sample efficient evaluation function in Chess Engines?” - as we integrate our evaluation function into modern Chess engines and play against agents with traditional Chess move evaluation functions, beating both random agents and a DeepChess implementation at a level-one search depth - representing the number of moves a traditional Chess agent (employing the alpha-beta search algorithm) looks ahead in order to evaluate a given chess state.
Tasks	Decision Making, Sentiment Analysis
Published	2019-07-18
URL	https://arxiv.org/abs/1907.08321v3
PDF	https://arxiv.org/pdf/1907.08321v3.pdf
PWC	https://paperswithcode.com/paper/sentimate-learning-to-play-chess-through
Repo
Framework