January 26, 2020

2959 words 14 mins read

Paper Group ANR 1545

PyramidBox++: High Performance Detector for Finding Tiny Face. Random Matrix Improved Covariance Estimation for a Large Class of Metrics. Further results on structured regression for multi-scale networks. Data Efficient Reinforcement Learning for Legged Robots. Implicit Recursive Characteristics of STOP. Machine Learning based English Sentiment Ana …

PyramidBox++: High Performance Detector for Finding Tiny Face


Title	PyramidBox++: High Performance Detector for Finding Tiny Face
Authors	Zhihang Li, Xu Tang, Junyu Han, Jingtuo Liu, Ran He
Abstract	With the rapid development of deep convolutional neural network, face detection has made great progress in recent years. WIDER FACE dataset, as a main benchmark, contributes greatly to this area. A large amount of methods have been put forward where PyramidBox designs an effective data augmentation strategy (Data-anchor-sampling) and context-based module for face detector. In this report, we improve each part to further boost the performance, including Balanced-data-anchor-sampling, Dual-PyramidAnchors and Dense Context Module. Specifically, Balanced-data-anchor-sampling obtains more uniform sampling of faces with different sizes. Dual-PyramidAnchors facilitate feature learning by introducing progressive anchor loss. Dense Context Module with dense connection not only enlarges receptive filed, but also passes information efficiently. Integrating these techniques, PyramidBox++ is constructed and achieves state-of-the-art performance in hard set.
Tasks	Data Augmentation, Face Detection
Published	2019-03-31
URL	http://arxiv.org/abs/1904.00386v1
PDF	http://arxiv.org/pdf/1904.00386v1.pdf
PWC	https://paperswithcode.com/paper/pyramidbox-high-performance-detector-for
Repo
Framework

Random Matrix Improved Covariance Estimation for a Large Class of Metrics


Title	Random Matrix Improved Covariance Estimation for a Large Class of Metrics
Authors	Malik Tiomoko, Florent Bouchard, Guillaume Ginholac, Romain Couillet
Abstract	Relying on recent advances in statistical estimation of covariance distances based on random matrix theory, this article proposes an improved covariance and precision matrix estimation for a wide family of metrics. The method is shown to largely outperform the sample covariance matrix estimate and to compete with state-of-the-art methods, while at the same time being computationally simpler. Applications to linear and quadratic discriminant analyses also demonstrate significant gains, therefore suggesting practical interest to statistical machine learning.
Tasks
Published	2019-02-07
URL	http://arxiv.org/abs/1902.02554v1
PDF	http://arxiv.org/pdf/1902.02554v1.pdf
PWC	https://paperswithcode.com/paper/random-matrix-improved-covariance-estimation
Repo
Framework

Further results on structured regression for multi-scale networks


Title	Further results on structured regression for multi-scale networks
Authors	Milan Bašić, Branko Arsić, Zoran Obradović
Abstract	Gaussian Conditional Random Fields (GCRF), as a structured regression model, is designed to achieve higher regression accuracy than unstructured predictors at the expense of execution time, taking into account the objects similarities and the outputs of unstructured predictors simultaneously. As most structural models, the GCRF model does not scale well with large networks. One of the approaches consists of performing calculations on factor graphs (if it is possible) rather than on the full graph, which is more computationally efficient. The Kronecker product of the graphs appears to be a natural choice for a graph decomposition. However, this idea is not straightforwardly applicable for GCRF, since characterizing a Laplacian spectrum of the Kronecker product of graphs, which GCRF is based on, from spectra of its factor graphs has remained an open problem. In this paper we apply new estimations for the Laplacian eigenvalues and eigenvectors, and achieve high prediction accuracy of the proposed models, while the computational complexity of the models, compared to the original GCRF model, is improved from $O(n_{1}^{3}n_{2}^{3})$ to $O(n_{1}^{3} + n_{2}^{3})$. Furthermore, we study the GCRF model with a non-Kronecker graph, where the model consists of finding the nearest Kronecker product of graph for an initial graph. Although the proposed models are more complex, they achieve high prediction accuracy too, while the execution time is still much better compare to the original GCRF model. The effectiveness of the proposed models is characterized on three types of random networks where the proposed models were consistently away more accurate than the previously presented GCRF model for multiscale networks [Jesse Glass and Zoran Obradovic. Structured regression on multiscale networks. IEEE Intelligent Systems, 32(2):23-30, 2017.].
Tasks
Published	2019-09-02
URL	https://arxiv.org/abs/1909.00853v1
PDF	https://arxiv.org/pdf/1909.00853v1.pdf
PWC	https://paperswithcode.com/paper/further-results-on-structured-regression-for
Repo
Framework

Data Efficient Reinforcement Learning for Legged Robots


Title	Data Efficient Reinforcement Learning for Legged Robots
Authors	Yuxiang Yang, Ken Caluwaerts, Atil Iscen, Tingnan Zhang, Jie Tan, Vikas Sindhwani
Abstract	We present a model-based framework for robot locomotion that achieves walking based on only 4.5 minutes (45,000 control steps) of data collected on a quadruped robot. To accurately model the robot’s dynamics over a long horizon, we introduce a loss function that tracks the model’s prediction over multiple timesteps. We adapt model predictive control to account for planning latency, which allows the learned model to be used for real time control. Additionally, to ensure safe exploration during model learning, we embed prior knowledge of leg trajectories into the action space. The resulting system achieves fast and robust locomotion. Unlike model-free methods, which optimize for a particular task, our planner can use the same learned dynamics for various tasks, simply by changing the reward function. To the best of our knowledge, our approach is more than an order of magnitude more sample efficient than current model-free methods.
Tasks	Legged Robots, Safe Exploration
Published	2019-07-08
URL	https://arxiv.org/abs/1907.03613v2
PDF	https://arxiv.org/pdf/1907.03613v2.pdf
PWC	https://paperswithcode.com/paper/data-efficient-reinforcement-learning-for
Repo
Framework

Implicit Recursive Characteristics of STOP


Title	Implicit Recursive Characteristics of STOP
Authors	Mike H. Ji
Abstract	The most important notations of Communicating Sequential Process(CSP) are the process and the prefix (event)$\rightarrow$(process) operator. While we can formally apply the $\rightarrow$ operator to define a live process’s behavior, the STOP process, which usually resulted from deadlock, starving or livelock, is lack of formal description, defined by most literatures as “doing nothing but halt”. In this paper, we argue that the STOP process should not be considered as a black box, it should follow the prefix $\rightarrow$ schema and the same inference rules so that a unified and consistent process algebra model can be established. In order to achieve this goal, we introduce a special event called “nil” that any process can take. This nil event will do nothing meaningful and leave nothing on a process’s observable record. With the nil event and its well-defined rules, we can successfully use the $\rightarrow$ operator to formally describe a process’s complete behavior in its whole life circle. More interestingly, we can use prefix $\rightarrow$ and nil event to fully describe the STOP process’s internal behavior and conclude that the STOP’s formal equation can be given as simple as STOP$_{\alpha X} = \mu$ X. nil $\rightarrow$ X.
Tasks
Published	2019-08-19
URL	https://arxiv.org/abs/1908.06601v2
PDF	https://arxiv.org/pdf/1908.06601v2.pdf
PWC	https://paperswithcode.com/paper/implicit-recursive-characteristics-of-stop
Repo
Framework

Machine Learning based English Sentiment Analysis


Title	Machine Learning based English Sentiment Analysis
Authors	T. N. T. Tran, L. K. N. Nguyen, V. M. Ngo
Abstract	Sentiment analysis or opinion mining aims to determine attitudes, judgments and opinions of customers for a product or a service. This is a great system to help manufacturers or servicers know the satisfaction level of customers about their products or services. From that, they can have appropriate adjustments. We use a popular machine learning method, being Support Vector Machine, combine with the library in Waikato Environment for Knowledge Analysis (WEKA) to build Java web program which analyzes the sentiment of English comments belongs one in four types of woman products. That are dresses, handbags, shoes and rings. We have developed and test our system with a training set having 300 comments and a test set having 400 comments. The experimental results of the system about precision, recall and F measures for positive comments are 89.3%, 95.0% and 92,.1%; for negative comments are 97.1%, 78.5% and 86.8%; and for neutral comments are 76.7%, 86.2% and 81.2%.
Tasks	Opinion Mining, Sentiment Analysis
Published	2019-05-16
URL	https://arxiv.org/abs/1905.06643v1
PDF	https://arxiv.org/pdf/1905.06643v1.pdf
PWC	https://paperswithcode.com/paper/machine-learning-based-english-sentiment
Repo
Framework

Illumination-invariant Face recognition by fusing thermal and visual images via gradient transfer


Title	Illumination-invariant Face recognition by fusing thermal and visual images via gradient transfer
Authors	Sumit Agarwal, Harshit S. Sikchi, Suparna Rooj, Shubhobrata Bhattacharya, Aurobinda Routray
Abstract	Face recognition in real life situations like low illumination condition is still an open challenge in biometric security. It is well established that the state-of-the-art methods in face recognition provide low accuracy in the case of poor illumination. In this work, we propose an algorithm for a more robust illumination invariant face recognition using a multi-modal approach. We propose a new dataset consisting of aligned faces of thermal and visual images of a hundred subjects. We then apply face detection on thermal images using the biggest blob extraction method and apply them for fusing images of different modalities for the purpose of face recognition. An algorithm is proposed to implement fusion of thermal and visual images. We reason for why relying on only one modality can give erroneous results. We use a lighter and faster CNN model called MobileNet for the purpose of face recognition with faster inferencing and to be able to be use it in real time biometric systems. We test our proposed method on our own created dataset to show that real-time face recognition on fused images shows far better results than using visual or thermal images separately.
Tasks	Face Detection, Face Recognition
Published	2019-02-23
URL	http://arxiv.org/abs/1902.08802v1
PDF	http://arxiv.org/pdf/1902.08802v1.pdf
PWC	https://paperswithcode.com/paper/illumination-invariant-face-recognition-by
Repo
Framework

Using Residual Dipolar Couplings from Two Alignment Media to Detect Structural Homology


Title	Using Residual Dipolar Couplings from Two Alignment Media to Detect Structural Homology
Authors	Ryan Yandle, Rishi Mukhopadhyay, Homayoun Valafar
Abstract	The method of Probability Density Profile Analysis has been introduced previously as a tool to find the best match between a set of experimentally generated Residual Dipolar Couplings and a set of known protein structures. While it proved effective on small databases in identifying protein fold families, and for picking the best result from computational protein folding tool ROBETTA, for larger data sets, more data is required. Here, the method of 2-D Probability Density Profile Analysis is presented which incorporates paired RDC data from 2 alignment media for N-H vectors. The method was tested using synthetic RDC data generated with +/-1 Hz error. The results show that the addition of information from a second alignment medium makes 2-D PDPA a much more effective tool that is able to identify a structure from a database of 600 protein fold family representatives.
Tasks
Published	2019-11-06
URL	https://arxiv.org/abs/1911.02396v1
PDF	https://arxiv.org/pdf/1911.02396v1.pdf
PWC	https://paperswithcode.com/paper/using-residual-dipolar-couplings-from-two
Repo
Framework

That which we call private


Title	That which we call private
Authors	Úlfar Erlingsson, Ilya Mironov, Ananth Raghunathan, Shuang Song
Abstract	A casual reader of the study by Jayaraman and Evans in USENIX Security 2019 might conclude that “relaxed definitions of differential privacy” should be avoided, because they “increase the measured privacy leakage.” This note clarifies that their study is consistent with a different interpretation. Namely, that the “relaxed definitions” are strict improvements which can improve the epsilon upper-bound guarantees by orders-of-magnitude without changing the actual privacy loss. Practitioners should be careful not to equate real-world privacy with epsilon values, without consideration of their context.
Tasks
Published	2019-08-08
URL	https://arxiv.org/abs/1908.03566v1
PDF	https://arxiv.org/pdf/1908.03566v1.pdf
PWC	https://paperswithcode.com/paper/that-which-we-call-private
Repo
Framework

Impoved RPN for Single Targets Detection based on the Anchor Mask Net


Title	Impoved RPN for Single Targets Detection based on the Anchor Mask Net
Authors	Mingjie Li, Youqian Feng, Zhonghai Yin, Cheng Zhou, Fanghao Dong
Abstract	Common target detection is usually based on single frame images, which is vulnerable to affected by the similar targets in the image and not applicable to video images. In this paper , anchor mask is proposed to add the prior knowledge for target detection and an anchor mask net is designed to impove the RPN performance for single target detection. Tested in the VOT2016, the model perform better.
Tasks
Published	2019-06-18
URL	https://arxiv.org/abs/1906.07527v1
PDF	https://arxiv.org/pdf/1906.07527v1.pdf
PWC	https://paperswithcode.com/paper/impoved-rpn-for-single-targets-detection
Repo
Framework

Distributed Differentially Private Computation of Functions with Correlated Noise


Title	Distributed Differentially Private Computation of Functions with Correlated Noise
Authors	Hafiz Imtiaz, Jafar Mohammadi, Anand D. Sarwate
Abstract	Many applications of machine learning, such as human health research, involve processing private or sensitive information. Privacy concerns may impose significant hurdles to collaboration in scenarios where there are multiple sites holding data and the goal is to estimate properties jointly across all datasets. Differentially private decentralized algorithms can provide strong privacy guarantees. However, the accuracy of the joint estimates may be poor when the datasets at each site are small. This paper proposes a new framework, Correlation Assisted Private Estimation (CAPE), for designing privacy-preserving decentralized algorithms with better accuracy guarantees in an honest-but-curious model. CAPE can be used in conjunction with the functional mechanism for statistical and machine learning optimization problems. A tighter characterization of the functional mechanism is provided that allows CAPE to achieve the same performance as a centralized algorithm in the decentralized setting using all datasets. Empirical results on regression and neural network problems for both synthetic and real datasets show that differentially private methods can be competitive with non-private algorithms in many scenarios of interest.
Tasks
Published	2019-04-22
URL	https://arxiv.org/abs/1904.10059v2
PDF	https://arxiv.org/pdf/1904.10059v2.pdf
PWC	https://paperswithcode.com/paper/distributed-differentially-private
Repo
Framework

Multimodal Embeddings from Language Models


Title	Multimodal Embeddings from Language Models
Authors	Shao-Yen Tseng, Panayiotis Georgiou, Shrikanth Narayanan
Abstract	Word embeddings such as ELMo have recently been shown to model word semantics with greater efficacy through contextualized learning on large-scale language corpora, resulting in significant improvement in state of the art across many natural language tasks. In this work we integrate acoustic information into contextualized lexical embeddings through the addition of multimodal inputs to a pretrained bidirectional language model. The language model is trained on spoken language that includes text and audio modalities. The resulting representations from this model are multimodal and contain paralinguistic information which can modify word meanings and provide affective information. We show that these multimodal embeddings can be used to improve over previous state of the art multimodal models in emotion recognition on the CMU-MOSEI dataset.
Tasks	Emotion Recognition, Language Modelling, Word Embeddings
Published	2019-09-10
URL	https://arxiv.org/abs/1909.04302v1
PDF	https://arxiv.org/pdf/1909.04302v1.pdf
PWC	https://paperswithcode.com/paper/multimodal-embeddings-from-language-models
Repo
Framework

Learning a Multitask Curriculum for Neural Machine Translation


Title	Learning a Multitask Curriculum for Neural Machine Translation
Authors	Wei Wang, Ye Tian, Jiquan Ngiam, Yinfei Yang, Isaac Caswell, Zarana Parekh
Abstract	Existing curriculum learning research in neural machine translation (NMT) mostly focuses on a single final task such as selecting data for a domain or for denoising, and considers in-task example selection. This paper studies the data selection problem in multitask setting. We present a method to learn a multitask curriculum on a single, diverse, potentially noisy training dataset. It computes multiple data selection scores for each training example, each score measuring how useful the example is to a certain task. It uses Bayesian optimization to learn a linear weighting of these per-instance scores, and then sorts the data to form a curriculum. We experiment with three domain translation tasks: two specific domains and the general domain, and demonstrate that the learned multitask curriculum delivers results close to individually optimized models and brings solid gains over no curriculum training, across all test sets.
Tasks	Denoising, Machine Translation
Published	2019-08-28
URL	https://arxiv.org/abs/1908.10940v1
PDF	https://arxiv.org/pdf/1908.10940v1.pdf
PWC	https://paperswithcode.com/paper/learning-a-multitask-curriculum-for-neural
Repo
Framework

Transfer Learning with Dynamic Adversarial Adaptation Network


Title	Transfer Learning with Dynamic Adversarial Adaptation Network
Authors	Chaohui Yu, Jindong Wang, Yiqiang Chen, Meiyu Huang
Abstract	The recent advances in deep transfer learning reveal that adversarial learning can be embedded into deep networks to learn more transferable features to reduce the distribution discrepancy between two domains. Existing adversarial domain adaptation methods either learn a single domain discriminator to align the global source and target distributions or pay attention to align subdomains based on multiple discriminators. However, in real applications, the marginal (global) and conditional (local) distributions between domains are often contributing differently to the adaptation. There is currently no method to dynamically and quantitatively evaluate the relative importance of these two distributions for adversarial learning. In this paper, we propose a novel Dynamic Adversarial Adaptation Network (DAAN) to dynamically learn domain-invariant representations while quantitatively evaluate the relative importance of global and local domain distributions. To the best of our knowledge, DAAN is the first attempt to perform dynamic adversarial distribution adaptation for deep adversarial learning. DAAN is extremely easy to implement and train in real applications. We theoretically analyze the effectiveness of DAAN, and it can also be explained in an attention strategy. Extensive experiments demonstrate that DAAN achieves better classification accuracy compared to state-of-the-art deep and adversarial methods. Results also imply the necessity and effectiveness of the dynamic distribution adaptation in adversarial transfer learning.
Tasks	Domain Adaptation, Transfer Learning
Published	2019-09-18
URL	https://arxiv.org/abs/1909.08184v1
PDF	https://arxiv.org/pdf/1909.08184v1.pdf
PWC	https://paperswithcode.com/paper/transfer-learning-with-dynamic-adversarial
Repo
Framework

A Principled Approach for Learning Task Similarity in Multitask Learning


Title	A Principled Approach for Learning Task Similarity in Multitask Learning
Authors	Changjian Shui, Mahdieh Abbasi, Louis-Émile Robitaille, Boyu Wang, Christian Gagné
Abstract	Multitask learning aims at solving a set of related tasks simultaneously, by exploiting the shared knowledge for improving the performance on individual tasks. Hence, an important aspect of multitask learning is to understand the similarities within a set of tasks. Previous works have incorporated this similarity information explicitly (e.g., weighted loss for each task) or implicitly (e.g., adversarial loss for feature adaptation), for achieving good empirical performances. However, the theoretical motivations for adding task similarity knowledge are often missing or incomplete. In this paper, we give a different perspective from a theoretical point of view to understand this practice. We first provide an upper bound on the generalization error of multitask learning, showing the benefit of explicit and implicit task similarity knowledge. We systematically derive the bounds based on two distinct task similarity metrics: H divergence and Wasserstein distance. From these theoretical results, we revisit the Adversarial Multi-task Neural Network, proposing a new training algorithm to learn the task relation coefficients and neural network parameters iteratively. We assess our new algorithm empirically on several benchmarks, showing not only that we find interesting and robust task relations, but that the proposed approach outperforms the baselines, reaffirming the benefits of theoretical insight in algorithm design.
Tasks
Published	2019-03-21
URL	https://arxiv.org/abs/1903.09109v2
PDF	https://arxiv.org/pdf/1903.09109v2.pdf
PWC	https://paperswithcode.com/paper/a-principled-approach-for-learning-task
Repo
Framework