October 20, 2019

3493 words 17 mins read

Paper Group ANR 72

Optimal stopping via reinforced regression. CNN-based Action Recognition and Supervised Domain Adaptation on 3D Body Skeletons via Kernel Feature Maps. Object Detection with Mask-based Feature Encoding. Deep Network Regularization via Bayesian Inference of Synaptic Connectivity. Pigeonring: A Principle for Faster Thresholded Similarity Search. IM2H …

Optimal stopping via reinforced regression


Title	Optimal stopping via reinforced regression
Authors	Denis Belomestny, John Schoenmakers, Vladimir Spokoiny, Bakhyt Zharkynbay
Abstract	In this note we propose a new approach towards solving numerically optimal stopping problems via reinforced regression based Monte Carlo algorithms. The main idea of the method is to reinforce standard linear regression algorithms in each backward induction step by adding new basis functions based on previously estimated continuation values. The proposed methodology is illustrated by a numerical example from mathematical finance.
Tasks
Published	2018-08-07
URL	https://arxiv.org/abs/1808.02341v3
PDF	https://arxiv.org/pdf/1808.02341v3.pdf
PWC	https://paperswithcode.com/paper/optimal-stopping-via-reinforced-regression
Repo
Framework

CNN-based Action Recognition and Supervised Domain Adaptation on 3D Body Skeletons via Kernel Feature Maps


Title	CNN-based Action Recognition and Supervised Domain Adaptation on 3D Body Skeletons via Kernel Feature Maps
Authors	Yusuf Tas, Piotr Koniusz
Abstract	Deep learning is ubiquitous across many areas areas of computer vision. It often requires large scale datasets for training before being fine-tuned on small-to-medium scale problems. Activity, or, in other words, action recognition, is one of many application areas of deep learning. While there exist many Convolutional Neural Network architectures that work with the RGB and optical flow frames, training on the time sequences of 3D body skeleton joints is often performed via recurrent networks such as LSTM. In this paper, we propose a new representation which encodes sequences of 3D body skeleton joints in texture-like representations derived from mathematically rigorous kernel methods. Such a representation becomes the first layer in a standard CNN network e.g., ResNet-50, which is then used in the supervised domain adaptation pipeline to transfer information from the source to target dataset. This lets us leverage the available Kinect-based data beyond training on a single dataset and outperform simple fine-tuning on any two datasets combined in a naive manner. More specifically, in this paper we utilize the overlapping classes between datasets. We associate datapoints of the same class via so-called commonality, known from the supervised domain adaptation. We demonstrate state-of-the-art results on three publicly available benchmarks.
Tasks	Domain Adaptation, Optical Flow Estimation, Temporal Action Localization
Published	2018-06-24
URL	http://arxiv.org/abs/1806.09078v1
PDF	http://arxiv.org/pdf/1806.09078v1.pdf
PWC	https://paperswithcode.com/paper/cnn-based-action-recognition-and-supervised
Repo
Framework

Object Detection with Mask-based Feature Encoding


Title	Object Detection with Mask-based Feature Encoding
Authors	Xiaochuan Fan, Hao Guo, Kang Zheng, Wei Feng, Song Wang
Abstract	Region-based Convolutional Neural Networks (R-CNNs) have achieved great success in the field of object detection. The existing R-CNNs usually divide a Region-of-Interest (ROI) into grids, and then localize objects by utilizing the spatial information reflected by the relative position of each grid in the ROI. In this paper, we propose a novel feature-encoding approach, where spatial information is represented through the spatial distributions of visual patterns. In particular, we design a Mask Weight Network (MWN) to learn a set of masks and then apply channel-wise masking operations to ROI feature map, followed by a global pooling and a cheap fully-connected layer. We integrate the newly designed feature encoder into the Faster R-CNN architecture. The resulting new Faster R-CNNs can preserve the object-detection accuracy of the standard Faster R-CNNs by using substantially fewer parameters. Compared to R-FCNs using state-of-art PS ROI pooling and deformable PS ROI pooling, the new Faster R-CNNs can produce higher object-detection accuracy with good run-time efficiency. We also show that a specifically designed and learned MWN can capture global contextual information and further improve the object-detection accuracy. Validation experiments are conducted on both PASCAL VOC and MS COCO datasets.
Tasks	Object Detection
Published	2018-02-12
URL	http://arxiv.org/abs/1802.03934v1
PDF	http://arxiv.org/pdf/1802.03934v1.pdf
PWC	https://paperswithcode.com/paper/object-detection-with-mask-based-feature
Repo
Framework

Deep Network Regularization via Bayesian Inference of Synaptic Connectivity


Title	Deep Network Regularization via Bayesian Inference of Synaptic Connectivity
Authors	Harris Partaourides, Sotirios P. Chatzis
Abstract	Deep neural networks (DNNs) often require good regularizers to generalize well. Currently, state-of-the-art DNN regularization techniques consist in randomly dropping units and/or connections on each iteration of the training algorithm. Dropout and DropConnect are characteristic examples of such regularizers, that are widely popular among practitioners. However, a drawback of such approaches consists in the fact that their postulated probability of random unit/connection omission is a constant that must be heuristically selected based on the obtained performance in some validation set. To alleviate this burden, in this paper we regard the DNN regularization problem from a Bayesian inference perspective: We impose a sparsity-inducing prior over the network synaptic weights, where the sparsity is induced by a set of Bernoulli-distributed binary variables with Beta (hyper-)priors over their prior parameters. This way, we eventually allow for marginalizing over the DNN synaptic connectivity for output generation, thus giving rise to an effective, heuristics-free, network regularization scheme. We perform Bayesian inference for the resulting hierarchical model by means of an efficient Black-Box Variational inference scheme. We exhibit the advantages of our method over existing approaches by conducting an extensive experimental evaluation using benchmark datasets.
Tasks	Bayesian Inference
Published	2018-03-04
URL	http://arxiv.org/abs/1803.01349v1
PDF	http://arxiv.org/pdf/1803.01349v1.pdf
PWC	https://paperswithcode.com/paper/deep-network-regularization-via-bayesian
Repo
Framework

Pigeonring: A Principle for Faster Thresholded Similarity Search


Title	Pigeonring: A Principle for Faster Thresholded Similarity Search
Authors	Jianbin Qin, Chuan Xiao
Abstract	The pigeonhole principle states that if $n$ items are contained in $m$ boxes, then at least one box has no more than $n / m$ items. It is utilized to solve many data management problems, especially for thresholded similarity searches. Despite many pigeonhole principle-based solutions proposed in the last few decades, the condition stated by the principle is weak. It only constrains the number of items in a single box. By organizing the boxes in a ring, we propose a new principle, called the pigeonring principle, which constrains the number of items in multiple boxes and yields stronger conditions. To utilize the new principle, we focus on problems defined in the form of identifying data objects whose similarities or distances to the query is constrained by a threshold. Many solutions to these problems utilize the pigeonhole principle to find candidates that satisfy a filtering condition. By the new principle, stronger filtering conditions can be established. We show that the pigeonhole principle is a special case of the new principle. This suggests that all the pigeonhole principle-based solutions are possible to be accelerated by the new principle. A universal filtering framework is introduced to encompass the solutions to these problems based on the new principle. Besides, we discuss how to quickly find candidates specified by the new principle. The implementation requires only minor modifications on top of existing pigeonhole principle-based algorithms. Experimental results on real datasets demonstrate the applicability of the new principle as well as the superior performance of the algorithms based on the new principle.
Tasks
Published	2018-04-04
URL	https://arxiv.org/abs/1804.01614v3
PDF	https://arxiv.org/pdf/1804.01614v3.pdf
PWC	https://paperswithcode.com/paper/pigeonring-a-principle-for-faster-thresholded
Repo
Framework

IM2HEIGHT: Height Estimation from Single Monocular Imagery via Fully Residual Convolutional-Deconvolutional Network


Title	IM2HEIGHT: Height Estimation from Single Monocular Imagery via Fully Residual Convolutional-Deconvolutional Network
Authors	Lichao Mou, Xiao Xiang Zhu
Abstract	In this paper we tackle a very novel problem, namely height estimation from a single monocular remote sensing image, which is inherently ambiguous, and a technically ill-posed problem, with a large source of uncertainty coming from the overall scale. We propose a fully convolutional-deconvolutional network architecture being trained end-to-end, encompassing residual learning, to model the ambiguous mapping between monocular remote sensing images and height maps. Specifically, it is composed of two parts, i.e., convolutional sub-network and deconvolutional sub-network. The former corresponds to feature extractor that transforms the input remote sensing image to high-level multidimensional feature representation, whereas the latter plays the role of a height generator that produces height map from the feature extracted from the convolutional sub-network. Moreover, to preserve fine edge details of estimated height maps, we introduce a skip connection to the network, which is able to shuttle low-level visual information, e.g., object boundaries and edges, directly across the network. To demonstrate the usefulness of single-view height prediction, we show a practical example of instance segmentation of buildings using estimated height map. This paper, for the first time in the remote sensing community, attempts to estimate height from monocular vision. The proposed network is validated using a large-scale high resolution aerial image data set covered an area of Berlin. Both visual and quantitative analysis of the experimental results demonstrate the effectiveness of our approach.
Tasks	Instance Segmentation, Semantic Segmentation
Published	2018-02-28
URL	http://arxiv.org/abs/1802.10249v1
PDF	http://arxiv.org/pdf/1802.10249v1.pdf
PWC	https://paperswithcode.com/paper/im2height-height-estimation-from-single
Repo
Framework

Multicolumn Networks for Face Recognition


Title	Multicolumn Networks for Face Recognition
Authors	Weidi Xie, Andrew Zisserman
Abstract	The objective of this work is set-based face recognition, i.e. to decide if two sets of images of a face are of the same person or not. Conventionally, the set-wise feature descriptor is computed as an average of the descriptors from individual face images within the set. In this paper, we design a neural network architecture that learns to aggregate based on both “visual” quality (resolution, illumination), and “content” quality (relative importance for discriminative classification). To this end, we propose a Multicolumn Network (MN) that takes a set of images (the number in the set can vary) as input, and learns to compute a fix-sized feature descriptor for the entire set. To encourage high-quality representations, each individual input image is first weighted by its “visual” quality, determined by a self-quality assessment module, and followed by a dynamic recalibration based on “content” qualities relative to the other images within the set. Both of these qualities are learnt implicitly during training for set-wise classification. Comparing with the previous state-of-the-art architectures trained with the same dataset (VGGFace2), our Multicolumn Networks show an improvement of between 2-6% on the IARPA IJB face recognition benchmarks, and exceed the state of the art for all methods on these benchmarks.
Tasks	Face Recognition
Published	2018-07-24
URL	http://arxiv.org/abs/1807.09192v1
PDF	http://arxiv.org/pdf/1807.09192v1.pdf
PWC	https://paperswithcode.com/paper/multicolumn-networks-for-face-recognition
Repo
Framework

Towards Understanding Acceleration Tradeoff between Momentum and Asynchrony in Nonconvex Stochastic Optimization


Title	Towards Understanding Acceleration Tradeoff between Momentum and Asynchrony in Nonconvex Stochastic Optimization
Authors	Tianyi Liu, Shiyang Li, Jianping Shi, Enlu Zhou, Tuo Zhao
Abstract	Asynchronous momentum stochastic gradient descent algorithms (Async-MSGD) is one of the most popular algorithms in distributed machine learning. However, its convergence properties for these complicated nonconvex problems is still largely unknown, because of the current technical limit. Therefore, in this paper, we propose to analyze the algorithm through a simpler but nontrivial nonconvex problem - streaming PCA, which helps us to understand Aync-MSGD better even for more general problems. Specifically, we establish the asymptotic rate of convergence of Async-MSGD for streaming PCA by diffusion approximation. Our results indicate a fundamental tradeoff between asynchrony and momentum: To ensure convergence and acceleration through asynchrony, we have to reduce the momentum (compared with Sync-MSGD). To the best of our knowledge, this is the first theoretical attempt on understanding Async-MSGD for distributed nonconvex stochastic optimization. Numerical experiments on both streaming PCA and training deep neural networks are provided to support our findings for Async-MSGD.
Tasks	Stochastic Optimization
Published	2018-06-04
URL	https://arxiv.org/abs/1806.01660v5
PDF	https://arxiv.org/pdf/1806.01660v5.pdf
PWC	https://paperswithcode.com/paper/towards-understanding-acceleration-tradeoff
Repo
Framework

Discovering Molecular Functional Groups Using Graph Convolutional Neural Networks


Title	Discovering Molecular Functional Groups Using Graph Convolutional Neural Networks
Authors	Phillip Pope, Soheil Kolouri, Mohammad Rostrami, Charles Martin, Heiko Hoffmann
Abstract	Functional groups (FGs) are molecular substructures that are served as a foundation for analyzing and predicting chemical properties of molecules. Automatic discovery of FGs will impact various fields of research, including medicinal chemistry and material sciences, by reducing the amount of lab experiments required for discovery or synthesis of new molecules. In this paper, we investigate methods based on graph convolutional neural networks (GCNNs) for localizing FGs that contribute to specific chemical properties of interest. In our framework, molecules are modeled as undirected relational graphs with atoms as nodes and bonds as edges. Using this relational graph structure, we trained GCNNs in a supervised way on experimentally-validated molecular training sets to predict specific chemical properties, e.g., toxicity. Upon learning a GCNN, we analyzed its activation patterns to automatically identify FGs using four different explainability methods that we have developed: gradient-based saliency maps, Class Activation Mapping (CAM), gradient-weighted CAM (Grad-CAM), and Excitation Back-Propagation. Although these methods are originally derived for convolutional neural networks (CNNs), we adapt them to develop the corresponding suitable versions for GCNNs. We evaluated the contrastive power of these methods with respect to the specificity of the identified molecular substructures and their relevance for chemical functions. Grad-CAM had the highest contrastive power and generated qualitatively the best FGs. This work paves the way for automatic analysis and design of new molecules.
Tasks
Published	2018-12-01
URL	https://arxiv.org/abs/1812.00265v3
PDF	https://arxiv.org/pdf/1812.00265v3.pdf
PWC	https://paperswithcode.com/paper/discovering-molecular-functional-groups-using
Repo
Framework

Convolutional Self-Attention Network


Title	Convolutional Self-Attention Network
Authors	Baosong Yang, Longyue Wang, Derek F. Wong, Lidia S. Chao, Zhaopeng Tu
Abstract	Self-attention network (SAN) has recently attracted increasing interest due to its fully parallelized computation and flexibility in modeling dependencies. It can be further enhanced with multi-headed attention mechanism by allowing the model to jointly attend to information from different representation subspaces at different positions (Vaswani et al., 2017). In this work, we propose a novel convolutional self-attention network (CSAN), which offers SAN the abilities to 1) capture neighboring dependencies, and 2) model the interaction between multiple attention heads. Experimental results on WMT14 English-to-German translation task demonstrate that the proposed approach outperforms both the strong Transformer baseline and other existing works on enhancing the locality of SAN. Comparing with previous work, our model does not introduce any new parameters.
Tasks
Published	2018-10-31
URL	http://arxiv.org/abs/1810.13320v2
PDF	http://arxiv.org/pdf/1810.13320v2.pdf
PWC	https://paperswithcode.com/paper/convolutional-self-attention-network
Repo
Framework

A Lyapunov-based Approach to Safe Reinforcement Learning


Title	A Lyapunov-based Approach to Safe Reinforcement Learning
Authors	Yinlam Chow, Ofir Nachum, Edgar Duenez-Guzman, Mohammad Ghavamzadeh
Abstract	In many real-world reinforcement learning (RL) problems, besides optimizing the main objective function, an agent must concurrently avoid violating a number of constraints. In particular, besides optimizing performance it is crucial to guarantee the safety of an agent during training as well as deployment (e.g. a robot should avoid taking actions - exploratory or not - which irrevocably harm its hardware). To incorporate safety in RL, we derive algorithms under the framework of constrained Markov decision problems (CMDPs), an extension of the standard Markov decision problems (MDPs) augmented with constraints on expected cumulative costs. Our approach hinges on a novel \emph{Lyapunov} method. We define and present a method for constructing Lyapunov functions, which provide an effective way to guarantee the global safety of a behavior policy during training via a set of local, linear constraints. Leveraging these theoretical underpinnings, we show how to use the Lyapunov approach to systematically transform dynamic programming (DP) and RL algorithms into their safe counterparts. To illustrate their effectiveness, we evaluate these algorithms in several CMDP planning and decision-making tasks on a safety benchmark domain. Our results show that our proposed method significantly outperforms existing baselines in balancing constraint satisfaction and performance.
Tasks	Decision Making
Published	2018-05-20
URL	http://arxiv.org/abs/1805.07708v1
PDF	http://arxiv.org/pdf/1805.07708v1.pdf
PWC	https://paperswithcode.com/paper/a-lyapunov-based-approach-to-safe
Repo
Framework

Twitter Sentiment Analysis System


Title	Twitter Sentiment Analysis System
Authors	Shaunak Joshi, Deepali Deshpande
Abstract	Social media is increasingly used by humans to express their feelings and opinions in the form of short text messages. Detecting sentiments in the text has a wide range of applications including identifying anxiety or depression of individuals and measuring well-being or mood of a community. Sentiments can be expressed in many ways that can be seen such as facial expression and gestures, speech and by written text. Sentiment Analysis in text documents is essentially a content-based classification problem involving concepts from the domains of Natural Language Processing as well as Machine Learning. In this paper, sentiment recognition based on textual data and the techniques used in sentiment analysis are discussed.
Tasks	Sentiment Analysis, Twitter Sentiment Analysis
Published	2018-07-20
URL	http://arxiv.org/abs/1807.07752v1
PDF	http://arxiv.org/pdf/1807.07752v1.pdf
PWC	https://paperswithcode.com/paper/twitter-sentiment-analysis-system
Repo
Framework

Chatter Classification in Turning Using Machine Learning and Topological Data Analysis


Title	Chatter Classification in Turning Using Machine Learning and Topological Data Analysis
Authors	Firas A. Khasawneh, Elizabeth Munch, Jose A. Perea
Abstract	Chatter identification and detection in machining processes has been an active area of research in the past two decades. Part of the challenge in studying chatter is that machining equations that describe its occurrence are often nonlinear delay differential equations. The majority of the available tools for chatter identification rely on defining a metric that captures the characteristics of chatter, and a threshold that signals its occurrence. The difficulty in choosing these parameters can be somewhat alleviated by utilizing machine learning techniques. However, even with a successful classification algorithm, the transferability of typical machine learning methods from one data set to another remains very limited. In this paper we combine supervised machine learning with Topological Data Analysis (TDA) to obtain a descriptor of the process which can detect chatter. The features we use are derived from the persistence diagram of an attractor reconstructed from the time series via Takens embedding. We test the approach using deterministic and stochastic turning models, where the stochasticity is introduced via the cutting coefficient term. Our results show a 97% successful classification rate on the deterministic model labeled by the stability diagram obtained using the spectral element method. The features gleaned from the deterministic model are then utilized for characterization of chatter in a stochastic turning model where there are very limited analysis methods.
Tasks	Time Series, Topological Data Analysis
Published	2018-03-23
URL	http://arxiv.org/abs/1804.02261v1
PDF	http://arxiv.org/pdf/1804.02261v1.pdf
PWC	https://paperswithcode.com/paper/chatter-classification-in-turning-using
Repo
Framework

Facing Multiple Attacks in Adversarial Patrolling Games with Alarmed Targets


Title	Facing Multiple Attacks in Adversarial Patrolling Games with Alarmed Targets
Authors	Giuseppe De Nittis, Nicola Gatti
Abstract	We focus on adversarial patrolling games on arbitrary graphs, where the Defender can control a mobile resource, the targets are alarmed by an alarm system, and the Attacker can observe the actions of the mobile resource of the Defender and perform different attacks exploiting multiple resources. This scenario can be modeled as a zero-sum extensive-form game in which each player can play multiple times. The game tree is exponentially large both in the size of the graph and in the number of attacking resources. We show that when the number of the Attacker’s resources is free, the problem of computing the equilibrium path is NP-hard, while when the number of resources is fixed, the equilibrium path can be computed in poly-time. We provide a dynamic-programming algorithm that, given the number of the Attacker’s resources, computes the equilibrium path requiring poly-time in the size of the graph and exponential time in the number of the resources. Furthermore, since in real-world scenarios it is implausible that the Defender knows the number of attacking resources, we study the robustness of the Defender’s strategy when she makes a wrong guess about that number. We show that even the error of just a single resource can lead to an arbitrary inefficiency, when the inefficiency is defined as the ratio of the Defender’s utilities obtained with a wrong guess and a correct guess. However, a more suitable definition of inefficiency is given by the difference of the Defender’s utilities: this way, we observe that the higher the error in the estimation, the higher the loss for the Defender. Then, we investigate the performance of online algorithms when no information about the Attacker’s resources is available. Finally, we resort to randomized online algorithms showing that we can obtain a competitive factor that is twice better than the one that can be achieved by any deterministic online algorithm.
Tasks
Published	2018-06-19
URL	http://arxiv.org/abs/1806.07111v1
PDF	http://arxiv.org/pdf/1806.07111v1.pdf
PWC	https://paperswithcode.com/paper/facing-multiple-attacks-in-adversarial
Repo
Framework

Probing Physics Knowledge Using Tools from Developmental Psychology


Title	Probing Physics Knowledge Using Tools from Developmental Psychology
Authors	Luis Piloto, Ari Weinstein, Dhruva TB, Arun Ahuja, Mehdi Mirza, Greg Wayne, David Amos, Chia-chun Hung, Matt Botvinick
Abstract	In order to build agents with a rich understanding of their environment, one key objective is to endow them with a grasp of intuitive physics; an ability to reason about three-dimensional objects, their dynamic interactions, and responses to forces. While some work on this problem has taken the approach of building in components such as ready-made physics engines, other research aims to extract general physical concepts directly from sensory data. In the latter case, one challenge that arises is evaluating the learning system. Research on intuitive physics knowledge in children has long employed a violation of expectations (VOE) method to assess children’s mastery of specific physical concepts. We take the novel step of applying this method to artificial learning systems. In addition to introducing the VOE technique, we describe a set of probe datasets inspired by classic test stimuli from developmental psychology. We test a baseline deep learning system on this battery, as well as on a physics learning dataset (“IntPhys”) recently posed by another research group. Our results show how the VOE technique may provide a useful tool for tracking physics knowledge in future research.
Tasks
Published	2018-04-03
URL	http://arxiv.org/abs/1804.01128v1
PDF	http://arxiv.org/pdf/1804.01128v1.pdf
PWC	https://paperswithcode.com/paper/probing-physics-knowledge-using-tools-from
Repo
Framework