October 20, 2019

3162 words 15 mins read

Paper Group AWR 297

Structured Control Nets for Deep Reinforcement Learning. Variational Bayesian Weighted Complex Network Reconstruction. TD or not TD: Analyzing the Role of Temporal Differencing in Deep Reinforcement Learning. RMDL: Random Multimodel Deep Learning for Classification. Multiview Supervision By Registration. Learning Dynamic Memory Networks for Object …

Structured Control Nets for Deep Reinforcement Learning


Title	Structured Control Nets for Deep Reinforcement Learning
Authors	Mario Srouji, Jian Zhang, Ruslan Salakhutdinov
Abstract	In recent years, Deep Reinforcement Learning has made impressive advances in solving several important benchmark problems for sequential decision making. Many control applications use a generic multilayer perceptron (MLP) for non-vision parts of the policy network. In this work, we propose a new neural network architecture for the policy network representation that is simple yet effective. The proposed Structured Control Net (SCN) splits the generic MLP into two separate sub-modules: a nonlinear control module and a linear control module. Intuitively, the nonlinear control is for forward-looking and global control, while the linear control stabilizes the local dynamics around the residual of global control. We hypothesize that this will bring together the benefits of both linear and nonlinear policies: improve training sample efficiency, final episodic reward, and generalization of learned policy, while requiring a smaller network and being generally applicable to different training methods. We validated our hypothesis with competitive results on simulations from OpenAI MuJoCo, Roboschool, Atari, and a custom 2D urban driving environment, with various ablation and generalization tests, trained with multiple black-box and policy gradient training methods. The proposed architecture has the potential to improve upon broader control tasks by incorporating problem specific priors into the architecture. As a case study, we demonstrate much improved performance for locomotion tasks by emulating the biological central pattern generators (CPGs) as the nonlinear part of the architecture.
Tasks	Decision Making
Published	2018-02-22
URL	http://arxiv.org/abs/1802.08311v1
PDF	http://arxiv.org/pdf/1802.08311v1.pdf
PWC	https://paperswithcode.com/paper/structured-control-nets-for-deep
Repo	https://github.com/wongongv/scnwithdqn
Framework	tf

Variational Bayesian Weighted Complex Network Reconstruction


Title	Variational Bayesian Weighted Complex Network Reconstruction
Authors	Shuang Xu, Chun-Xia Zhang, Pei Wang, Jiangshe Zhang
Abstract	Complex network reconstruction is a hot topic in many fields. Currently, the most popular data-driven reconstruction framework is based on lasso. However, it is found that, in the presence of noise, lasso loses efficiency for weighted networks. This paper builds a new framework to cope with this problem. The key idea is to employ a series of linear regression problems to model the relationship between network nodes, and then to use an efficient variational Bayesian algorithm to infer the unknown coefficients. The numerical experiments conducted on both synthetic and real data demonstrate that the new method outperforms lasso with regard to both reconstruction accuracy and running speed.
Tasks
Published	2018-12-11
URL	https://arxiv.org/abs/1812.04369v3
PDF	https://arxiv.org/pdf/1812.04369v3.pdf
PWC	https://paperswithcode.com/paper/variational-bayesian-complex-network
Repo	https://github.com/xsxjtu/VBR
Framework	none

TD or not TD: Analyzing the Role of Temporal Differencing in Deep Reinforcement Learning


Title	TD or not TD: Analyzing the Role of Temporal Differencing in Deep Reinforcement Learning
Authors	Artemij Amiranashvili, Alexey Dosovitskiy, Vladlen Koltun, Thomas Brox
Abstract	Our understanding of reinforcement learning (RL) has been shaped by theoretical and empirical results that were obtained decades ago using tabular representations and linear function approximators. These results suggest that RL methods that use temporal differencing (TD) are superior to direct Monte Carlo estimation (MC). How do these results hold up in deep RL, which deals with perceptually complex environments and deep nonlinear models? In this paper, we re-examine the role of TD in modern deep RL, using specially designed environments that control for specific factors that affect performance, such as reward sparsity, reward delay, and the perceptual complexity of the task. When comparing TD with infinite-horizon MC, we are able to reproduce classic results in modern settings. Yet we also find that finite-horizon MC is not inferior to TD, even when rewards are sparse or delayed. This makes MC a viable alternative to TD in deep RL.
Tasks
Published	2018-06-04
URL	http://arxiv.org/abs/1806.01175v1
PDF	http://arxiv.org/pdf/1806.01175v1.pdf
PWC	https://paperswithcode.com/paper/td-or-not-td-analyzing-the-role-of-temporal
Repo	https://github.com/lmb-freiburg/td-or-not-td
Framework	tf

RMDL: Random Multimodel Deep Learning for Classification


Title	RMDL: Random Multimodel Deep Learning for Classification
Authors	Kamran Kowsari, Mojtaba Heidarysafa, Donald E. Brown, Kiana Jafari Meimandi, Laura E. Barnes
Abstract	The continually increasing number of complex datasets each year necessitates ever improving machine learning methods for robust and accurate categorization of these data. This paper introduces Random Multimodel Deep Learning (RMDL): a new ensemble, deep learning approach for classification. Deep learning models have achieved state-of-the-art results across many domains. RMDL solves the problem of finding the best deep learning structure and architecture while simultaneously improving robustness and accuracy through ensembles of deep learning architectures. RDML can accept as input a variety data to include text, video, images, and symbolic. This paper describes RMDL and shows test results for image and text data including MNIST, CIFAR-10, WOS, Reuters, IMDB, and 20newsgroup. These test results show that RDML produces consistently better performance than standard methods over a broad range of data types and classification problems.
Tasks	Document Classification, Face Recognition, Image Classification, Multi-Label Text Classification
Published	2018-05-03
URL	http://arxiv.org/abs/1805.01890v2
PDF	http://arxiv.org/pdf/1805.01890v2.pdf
PWC	https://paperswithcode.com/paper/rmdl-random-multimodel-deep-learning-for
Repo	https://github.com/kk7nc/RMDL
Framework	tf

Multiview Supervision By Registration


Title	Multiview Supervision By Registration
Authors	Yilun Zhang, Hyun Soo Park
Abstract	This paper presents a semi-supervised learning framework to train a keypoint detector using multiview image streams given the limited labeled data (typically $<$4%). We leverage the complementary relationship between multiview geometry and visual tracking to provide three types of supervisionary signals to utilize the unlabeled data: (1) keypoint detection in one view can be supervised by other views via the epipolar geometry; (2) a keypoint moves smoothly over time where its optical flow can be used to temporally supervise consecutive image frames to each other; (3) visible keypoint in one view is likely to be visible in the adjacent view. We integrate these three signals in a differentiable fashion to design a new end-to-end neural network composed of three pathways. This design allows us to extensively use the unlabeled data to train the keypoint detector. We show that our approach outperforms existing detectors including DeepLabCut tailored to the keypoint detection of non-human species such as monkeys, dogs, and mice.
Tasks	3D Reconstruction, Keypoint Detection, Optical Flow Estimation, Visual Tracking
Published	2018-11-27
URL	http://arxiv.org/abs/1811.11251v2
PDF	http://arxiv.org/pdf/1811.11251v2.pdf
PWC	https://paperswithcode.com/paper/multiview-supervision-by-registration
Repo	https://github.com/msbrpp/MSBR
Framework	pytorch

Learning Dynamic Memory Networks for Object Tracking


Title	Learning Dynamic Memory Networks for Object Tracking
Authors	Tianyu Yang, Antoni B. Chan
Abstract	Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object’s appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic memory network to adapt the template to the target’s appearance variations during tracking. An LSTM is used as a memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the memory block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object’s information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target’s appearance changes by updating the external memory. Moreover, unlike other tracking methods where the model capacity is fixed after offline training — the capacity of our tracker can be easily enlarged as the memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.
Tasks	Object Tracking, Visual Tracking
Published	2018-03-20
URL	http://arxiv.org/abs/1803.07268v2
PDF	http://arxiv.org/pdf/1803.07268v2.pdf
PWC	https://paperswithcode.com/paper/learning-dynamic-memory-networks-for-object
Repo	https://github.com/skyoung/MemTrack
Framework	tf

Learning RoI Transformer for Detecting Oriented Objects in Aerial Images


Title	Learning RoI Transformer for Detecting Oriented Objects in Aerial Images
Authors	Jian Ding, Nan Xue, Yang Long, Gui-Song Xia, Qikai Lu
Abstract	Object detection in aerial images is an active yet challenging task in computer vision because of the birdview perspective, the highly complex backgrounds, and the variant appearances of objects. Especially when detecting densely packed objects in aerial images, methods relying on horizontal proposals for common object detection often introduce mismatches between the Region of Interests (RoIs) and objects. This leads to the common misalignment between the final object classification confidence and localization accuracy. Although rotated anchors have been used to tackle this problem, the design of them always multiplies the number of anchors and dramatically increases the computational complexity. In this paper, we propose a RoI Transformer to address these problems. More precisely, to improve the quality of region proposals, we first designed a Rotated RoI (RRoI) learner to transform a Horizontal Region of Interest (HRoI) into a Rotated Region of Interest (RRoI). Based on the RRoIs, we then proposed a Rotated Position Sensitive RoI Align (RPS-RoI-Align) module to extract rotation-invariant features from them for boosting subsequent classification and regression. Our RoI Transformer is with light weight and can be easily embedded into detectors for oriented object detection. A simple implementation of the RoI Transformer has achieved state-of-the-art performances on two common and challenging aerial datasets, i.e., DOTA and HRSC2016, with a neglectable reduction to detection speed. Our RoI Transformer exceeds the deformable Position Sensitive RoI pooling when oriented bounding-box annotations are available. Extensive experiments have also validated the flexibility and effectiveness of our RoI Transformer. The results demonstrate that it can be easily integrated with other detector architectures and significantly improve the performances.
Tasks	Object Classification, Object Detection, Object Detection In Aerial Images
Published	2018-12-01
URL	http://arxiv.org/abs/1812.00155v1
PDF	http://arxiv.org/pdf/1812.00155v1.pdf
PWC	https://paperswithcode.com/paper/learning-roi-transformer-for-detecting
Repo	https://github.com/dingjiansw101/RoITransformer_DOTA
Framework	mxnet

A Feature-Rich Vietnamese Named-Entity Recognition Model


Title	A Feature-Rich Vietnamese Named-Entity Recognition Model
Authors	Pham Quang Nhat Minh
Abstract	In this paper, we present a feature-based named-entity recognition (NER) model that achieves the start-of-the-art accuracy for Vietnamese language. We combine word, word-shape features, PoS, chunk, Brown-cluster-based features, and word-embedding-based features in the Conditional Random Fields (CRF) model. We also explore the effects of word segmentation, PoS tagging, and chunking results of many popular Vietnamese NLP toolkits on the accuracy of the proposed feature-based NER model. Up to now, our work is the first work that systematically performs an extrinsic evaluation of basic Vietnamese NLP toolkits on the downstream NER task. Experimental results show that while automatically-generated word segmentation is useful, PoS and chunking information generated by Vietnamese NLP tools does not show their benefits for the proposed feature-based NER model.
Tasks	Chunking, Named Entity Recognition
Published	2018-03-12
URL	http://arxiv.org/abs/1803.04375v1
PDF	http://arxiv.org/pdf/1803.04375v1.pdf
PWC	https://paperswithcode.com/paper/a-feature-rich-vietnamese-named-entity
Repo	https://github.com/minhpqn/vietner
Framework	none

IP Geolocation through Reverse DNS


Title	IP Geolocation through Reverse DNS
Authors	Ovidiu Dan, Vaibhav Parikh, Brian D. Davison
Abstract	IP Geolocation databases are widely used in online services to map end user IP addresses to their geographical locations. However, they use proprietary geolocation methods and in some cases they have poor accuracy. We propose a systematic approach to use publicly accessible reverse DNS hostnames for geolocating IP addresses. Our method is designed to be combined with other geolocation data sources. We cast the task as a machine learning problem where for a given hostname, we generate and rank a list of potential location candidates. We evaluate our approach against three state of the art academic baselines and two state of the art commercial IP geolocation databases. We show that our work significantly outperforms the academic baselines, and is complementary and competitive with commercial databases. To aid reproducibility, we open source our entire approach.
Tasks
Published	2018-11-10
URL	http://arxiv.org/abs/1811.04288v1
PDF	http://arxiv.org/pdf/1811.04288v1.pdf
PWC	https://paperswithcode.com/paper/ip-geolocation-through-reverse-dns
Repo	https://github.com/furiousxk/Docs2Go
Framework	none

Multimodal Grounding for Language Processing


Title	Multimodal Grounding for Language Processing
Authors	Lisa Beinborn, Teresa Botschen, Iryna Gurevych
Abstract	This survey discusses how recent developments in multimodal processing facilitate conceptual grounding of language. We categorize the information flow in multimodal processing with respect to cognitive models of human information processing and analyze different methods for combining multimodal representations. Based on this methodological inventory, we discuss the benefit of multimodal grounding for a variety of language processing tasks and the challenges that arise. We particularly focus on multimodal grounding of verbs which play a crucial role for the compositional power of language.
Tasks
Published	2018-06-17
URL	https://arxiv.org/abs/1806.06371v2
PDF	https://arxiv.org/pdf/1806.06371v2.pdf
PWC	https://paperswithcode.com/paper/multimodal-grounding-for-language-processing
Repo	https://github.com/UKPLab/coling18-multimodalSurvey
Framework	none

Variational Inference with Tail-adaptive f-Divergence


Title	Variational Inference with Tail-adaptive f-Divergence
Authors	Dilin Wang, Hao Liu, Qiang Liu
Abstract	Variational inference with {\alpha}-divergences has been widely used in modern probabilistic machine learning. Compared to Kullback-Leibler (KL) divergence, a major advantage of using {\alpha}-divergences (with positive {\alpha} values) is their mass-covering property. However, estimating and optimizing {\alpha}-divergences require to use importance sampling, which could have extremely large or infinite variances due to heavy tails of importance weights. In this paper, we propose a new class of tail-adaptive f-divergences that adaptively change the convex function f with the tail of the importance weights, in a way that theoretically guarantees finite moments, while simultaneously achieving mass-covering properties. We test our methods on Bayesian neural networks, as well as deep reinforcement learning in which our method is applied to improve a recent soft actor-critic (SAC) algorithm. Our results show that our approach yields significant advantages compared with existing methods based on classical KL and {\alpha}-divergences.
Tasks
Published	2018-10-29
URL	https://arxiv.org/abs/1810.11943v3
PDF	https://arxiv.org/pdf/1810.11943v3.pdf
PWC	https://paperswithcode.com/paper/variational-inference-with-tail-adaptive-f
Repo	https://github.com/dilinwang820/adaptive-f-divergence
Framework	tf

Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series


Title	Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series
Authors	Charlotte Pelletier, Geoffrey I. Webb, Francois Petitjean
Abstract	New remote sensing sensors now acquire high spatial and spectral Satellite Image Time Series (SITS) of the world. These series of images are a key component of classification systems that aim at obtaining up-to-date and accurate land cover maps of the Earth’s surfaces. More specifically, the combination of the temporal, spectral and spatial resolutions of new SITS makes possible to monitor vegetation dynamics. Although traditional classification algorithms, such as Random Forest (RF), have been successfully applied for SITS classification, these algorithms do not make the most of the temporal domain. Conversely, some approaches that take into account the temporal dimension have recently been tested, especially Recurrent Neural Networks (RNNs). This paper proposes an exhaustive study of another deep learning approaches, namely Temporal Convolutional Neural Networks (TempCNNs) where convolutions are applied in the temporal dimension. The goal is to quantitatively and qualitatively evaluate the contribution of TempCNNs for SITS classification. This paper proposes a set of experiments performed on one million time series extracted from 46 Formosat-2 images. The experimental results show that TempCNNs are more accurate than RF and RNNs, that are the current state of the art for SITS classification. We also highlight some differences with results obtained in computer vision, e.g. about pooling layers. Moreover, we provide some general guidelines on the network architecture, common regularization mechanisms, and hyper-parameter values such as batch size. Finally, we assess the visual quality of the land cover maps produced by TempCNNs.
Tasks	Image Classification, Time Series
Published	2018-11-26
URL	http://arxiv.org/abs/1811.10166v2
PDF	http://arxiv.org/pdf/1811.10166v2.pdf
PWC	https://paperswithcode.com/paper/temporal-convolutional-neural-network-for-the
Repo	https://github.com/charlotte-pel/temporalCNN
Framework	tf

IOHprofiler: A Benchmarking and Profiling Tool for Iterative Optimization Heuristics


Title	IOHprofiler: A Benchmarking and Profiling Tool for Iterative Optimization Heuristics
Authors	Carola Doerr, Hao Wang, Furong Ye, Sander van Rijn, Thomas Bäck
Abstract	IOHprofiler is a new tool for analyzing and comparing iterative optimization heuristics. Given as input algorithms and problems written in C or Python, it provides as output a statistical evaluation of the algorithms’ performance by means of the distribution on the fixed-target running time and the fixed-budget function values. In addition, IOHprofiler also allows to track the evolution of algorithm parameters, making our tool particularly useful for the analysis, comparison, and design of (self-)adaptive algorithms. IOHprofiler is a ready-to-use software. It consists of two parts: an experimental part, which generates the running time data, and a post-processing part, which produces the summarizing comparisons and statistical evaluations. The experimental part is build on the COCO software, which has been adjusted to cope with optimization problems that are formulated as functions $f:\mathcal{S}^n \to \R$ with $\mathcal{S}$ being a discrete alphabet of integers. The post-processing part is our own work. It can be used as a stand-alone tool for the evaluation of running time data of arbitrary benchmark problems. It accepts as input files not only the output files of IOHprofiler, but also original COCO data files. The post-processing tool is designed for an interactive evaluation, allowing the user to chose the ranges and the precision of the displayed data according to his/her needs. IOHprofiler is available on GitHub at \url{https://github.com/IOHprofiler}.
Tasks
Published	2018-10-11
URL	http://arxiv.org/abs/1810.05281v1
PDF	http://arxiv.org/pdf/1810.05281v1.pdf
PWC	https://paperswithcode.com/paper/iohprofiler-a-benchmarking-and-profiling-tool
Repo	https://github.com/IOHprofiler/IOHexperimenter
Framework	none

Real-Time Sleep Staging using Deep Learning on a Smartphone for a Wearable EEG


Title	Real-Time Sleep Staging using Deep Learning on a Smartphone for a Wearable EEG
Authors	Abhay Koushik, Judith Amores, Pattie Maes
Abstract	We present the first real-time sleep staging system that uses deep learning without the need for servers in a smartphone application for a wearable EEG. We employ real-time adaptation of a single channel Electroencephalography (EEG) to infer from a Time-Distributed 1-D Deep Convolutional Neural Network. Polysomnography (PSG)-the gold standard for sleep staging, requires a human scorer and is both complex and resource-intensive. Our work demonstrates an end-to-end on-smartphone pipeline that can infer sleep stages in just single 30-second epochs, with an overall accuracy of 83.5% on 20-fold cross validation for five-class classification of sleep stages using the open Sleep-EDF dataset.
Tasks	EEG, Sleep Stage Detection
Published	2018-11-25
URL	http://arxiv.org/abs/1811.10111v2
PDF	http://arxiv.org/pdf/1811.10111v2.pdf
PWC	https://paperswithcode.com/paper/real-time-sleep-staging-using-deep-learning
Repo	https://github.com/kylemath/DeepEEG
Framework	tf

Adversarial Example Generation with Syntactically Controlled Paraphrase Networks


Title	Adversarial Example Generation with Syntactically Controlled Paraphrase Networks
Authors	Mohit Iyyer, John Wieting, Kevin Gimpel, Luke Zettlemoyer
Abstract	We propose syntactically controlled paraphrase networks (SCPNs) and use them to generate adversarial examples. Given a sentence and a target syntactic form (e.g., a constituency parse), SCPNs are trained to produce a paraphrase of the sentence with the desired syntax. We show it is possible to create training data for this task by first doing backtranslation at a very large scale, and then using a parser to label the syntactic transformations that naturally occur during this process. Such data allows us to train a neural encoder-decoder model with extra inputs to specify the target syntax. A combination of automated and human evaluations show that SCPNs generate paraphrases that follow their target specifications without decreasing paraphrase quality when compared to baseline (uncontrolled) paraphrase systems. Furthermore, they are more capable of generating syntactically adversarial examples that both (1) “fool” pretrained models and (2) improve the robustness of these models to syntactic variation when used to augment their training data.
Tasks
Published	2018-04-17
URL	http://arxiv.org/abs/1804.06059v1
PDF	http://arxiv.org/pdf/1804.06059v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-example-generation-with
Repo	https://github.com/miyyer/scpn
Framework	pytorch