January 26, 2020

3176 words 15 mins read

Paper Group ANR 1372

Training Data Independent Image Registration With GANs Using Transfer Learning And Segmentation Information. Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning. A Neural Network-Based On-device Learning Anomaly Detector for Edge Devices. Improved robustness of reinforcement learning policies upon conversion to spiking …

Training Data Independent Image Registration With GANs Using Transfer Learning And Segmentation Information


Title	Training Data Independent Image Registration With GANs Using Transfer Learning And Segmentation Information
Authors	Dwarikanath Mahapatra, Zongyuan Ge
Abstract	Registration is an important task in automated medical image analysis. Although deep learning (DL) based image registration methods out perform time consuming conventional approaches, they are heavily dependent on training data and do not generalize well for new images types. We present a DL based approach that can register an image pair which is different from the training images. This is achieved by training generative adversarial networks (GANs) in combination with segmentation information and transfer learning. Experiments on chest Xray and brain MR images show that our method gives better registration performance over conventional methods.
Tasks	Image Registration, Transfer Learning
Published	2019-03-25
URL	http://arxiv.org/abs/1903.10139v2
PDF	http://arxiv.org/pdf/1903.10139v2.pdf
PWC	https://paperswithcode.com/paper/combining-transfer-learning-and-segmentation
Repo
Framework

Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning


Title	Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning
Authors	Peng Xu, Chien-Sheng Wu, Andrea Madotto, Pascale Fung
Abstract	Sensational headlines are headlines that capture people’s attention and generate reader interest. Conventional abstractive headline generation methods, unlike human writers, do not optimize for maximal reader attention. In this paper, we propose a model that generates sensational headlines without labeled data. We first train a sensationalism scorer by classifying online headlines with many comments (“clickbait”) against a baseline of headlines generated from a summarization model. The score from the sensationalism scorer is used as the reward for a reinforcement learner. However, maximizing the noisy sensationalism reward will generate unnatural phrases instead of sensational headlines. To effectively leverage this noisy reward, we propose a novel loss function, Auto-tuned Reinforcement Learning (ARL), to dynamically balance reinforcement learning (RL) with maximum likelihood estimation (MLE). Human evaluation shows that 60.8% of samples generated by our model are sensational, which is significantly better than the Pointer-Gen baseline and other RL models.
Tasks
Published	2019-09-09
URL	https://arxiv.org/abs/1909.03582v1
PDF	https://arxiv.org/pdf/1909.03582v1.pdf
PWC	https://paperswithcode.com/paper/clickbait-sensational-headline-generation
Repo
Framework

A Neural Network-Based On-device Learning Anomaly Detector for Edge Devices


Title	A Neural Network-Based On-device Learning Anomaly Detector for Edge Devices
Authors	Mineto Tsukada, Masaaki Kondo, Hiroki Matsutani
Abstract	Semi-supervised anomaly detection is an approach to identify anomalies by learning the distribution of normal data. Backpropagation neural networks (i.e., BP-NNs) based approaches have recently drawn attention because of their good generalization capability. In a typical situation, BP-NN-based models are iteratively optimized in server machines with input data gathered from edge devices. However, (1) the iterative optimization often requires significant efforts to follow changes in the distribution of normal data (i.e., concept drift), and (2) data transfers between edge and server impose additional latency and energy consumption. To address these issues, we propose ONLAD and its IP core, named ONLAD Core. ONLAD is highly optimized to perform fast sequential learning to follow concept drift in less than one millisecond. ONLAD Core realizes on-device learning for edge devices at low power consumption, which realizes standalone execution where data transfers between edge and server are not required. Experiments show that ONLAD has favorable anomaly detection capability in an environment that simulates concept drift. Evaluations of ONLAD Core confirm that the training latency is 1.95x~6.58x faster than the other software implementations. Also, the runtime power consumption of ONLAD Core implemented on PYNQ-Z1 board, a small FPGA/CPU SoC platform, is 5.0x~25.4x lower than them.
Tasks	Anomaly Detection, Time Series
Published	2019-07-23
URL	https://arxiv.org/abs/1907.10147v5
PDF	https://arxiv.org/pdf/1907.10147v5.pdf
PWC	https://paperswithcode.com/paper/a-neural-network-based-on-device-learning
Repo
Framework

Improved robustness of reinforcement learning policies upon conversion to spiking neuronal network platforms applied to ATARI games


Title	Improved robustness of reinforcement learning policies upon conversion to spiking neuronal network platforms applied to ATARI games
Authors	Devdhar Patel, Hananel Hazan, Daniel J. Saunders, Hava Siegelmann, Robert Kozma
Abstract	Deep Reinforcement Learning (RL) demonstrates excellent performance on tasks that can be solved by trained policy. It plays a dominant role among cutting-edge machine learning approaches using multi-layer Neural networks (NNs). At the same time, Deep RL suffers from high sensitivity to noisy, incomplete, and misleading input data. Following biological intuition, we involve Spiking Neural Networks (SNNs) to address some deficiencies of deep RL solutions. Previous studies in image classification domain demonstrated that standard NNs (with ReLU nonlinearity) trained using supervised learning can be converted to SNNs with negligible deterioration in performance. In this paper, we extend those conversion results to the domain of Q-Learning NNs trained using RL. We provide a proof of principle of the conversion of standard NN to SNN. In addition, we show that the SNN has improved robustness to occlusion in the input image. Finally, we introduce results with converting full-scale Deep Q-network to SNN, paving the way for future research to robust Deep RL applications.
Tasks	Atari Games, Image Classification, Q-Learning
Published	2019-03-26
URL	https://arxiv.org/abs/1903.11012v3
PDF	https://arxiv.org/pdf/1903.11012v3.pdf
PWC	https://paperswithcode.com/paper/improved-robustness-of-reinforcement-learning
Repo
Framework

From Caesar Cipher to Unsupervised Learning: A New Method for Classifier Parameter Estimation


Title	From Caesar Cipher to Unsupervised Learning: A New Method for Classifier Parameter Estimation
Authors	Yu Liu, Li Deng, Jianshu Chen, Chang Wen Chen
Abstract	Many important classification problems, such as object classification, speech recognition, and machine translation, have been tackled by the supervised learning paradigm in the past, where training corpora of parallel input-output pairs are required with high cost. To remove the need for the parallel training corpora has practical significance for real-world applications, and it is one of the main goals of unsupervised learning. Recently, encouraging progress in unsupervised learning for solving such classification problems has been made and the nature of the challenges has been clarified. In this article, we review this progress and disseminate a class of promising new methods to facilitate understanding the methods for machine learning researchers. In particular, we emphasize the key information that enables the success of unsupervised learning - the sequential statistics as the distributional prior in the labels. Exploitation of such sequential statistics makes it possible to estimate parameters of classifiers without the need of paired input-output data. In this paper, we first introduce the concept of Caesar Cipher and its decryption, which motivated the construction of the novel loss function for unsupervised learning we use throughout the paper. Then we use a simple but representative binary classification task as an example to derive and describe the unsupervised learning algorithm in a step-by-step, easy-to-understand fashion. We include two cases, one with Bigram language model as the sequential statistics for use in unsupervised parameter estimation, and another with a simpler Unigram language model. For both cases, detailed derivation steps for the learning algorithm are included. Further, a summary table compares computational steps of the two cases in executing the unsupervised learning algorithm for learning binary classifiers.
Tasks	Language Modelling, Machine Translation, Object Classification, Speech Recognition
Published	2019-06-06
URL	https://arxiv.org/abs/1906.02826v1
PDF	https://arxiv.org/pdf/1906.02826v1.pdf
PWC	https://paperswithcode.com/paper/from-caesar-cipher-to-unsupervised-learning-a
Repo
Framework

Farm land weed detection with region-based deep convolutional neural networks


Title	Farm land weed detection with region-based deep convolutional neural networks
Authors	Mohammad Ibrahim Sarker, Hyongsuk Kim
Abstract	Machine learning has become a major field of research in order to handle more and more complex image detection problems. Among the existing state-of-the-art CNN models, in this paper a region-based, fully convolutional network, for fast and accurate object detection has been proposed based on the experimental results. Among the region based networks, ResNet is regarded as the most recent CNN architecture which has obtained the best results at ImageNet Large-Scale Visual Recognition Challenge (ILSVRC) in 2015. Deep residual networks (ResNets) can make the training process faster and attain more accuracy compared to their equivalent conventional neural networks. Being motivated with such unique attributes of ResNet, this paper evaluates the performance of fine-tuned ResNet for object classification of our weeds dataset. The dataset of farm land weeds detection is insufficient to train such deep CNN models. To overcome this shortcoming, we perform dropout techniques along with deep residual network for reducing over-fitting problem as well as applying data augmentation with the proposed ResNet to achieve a significant outperforming result from our weeds dataset. We achieved better object detection performance with Region-based Fully Convolutional Networks (R-FCN) technique which is latched with our proposed ResNet-101.
Tasks	Data Augmentation, Object Classification, Object Detection, Object Recognition
Published	2019-06-05
URL	https://arxiv.org/abs/1906.01885v1
PDF	https://arxiv.org/pdf/1906.01885v1.pdf
PWC	https://paperswithcode.com/paper/farm-land-weed-detection-with-region-based
Repo
Framework

Neural Network-based Object Classification by Known and Unknown Features (Based on Text Queries)


Title	Neural Network-based Object Classification by Known and Unknown Features (Based on Text Queries)
Authors	A. Artemov, I. Bolokhov, D. Kem, I. Khasenevich
Abstract	The article presents a method that improves the quality of classification of objects described by a combination of known and unknown features. The method is based on modernized Informational Neurobayesian Approach with consideration of unknown features. The proposed method was developed and trained on 1500 text queries of Promobot users in Russian to classify them into 20 categories (classes). As a result, the use of the method allowed to completely solve the problem of misclassification for queries with combining known and unknown features of the model. The theoretical substantiation of the method is presented by the formulated and proved theorem On the Model with Limited Knowledge. It states, that in conditions of limited data, an equal number of equally unknown features of an object cannot have different significance for the classification problem.
Tasks	Object Classification
Published	2019-06-03
URL	https://arxiv.org/abs/1906.00800v1
PDF	https://arxiv.org/pdf/1906.00800v1.pdf
PWC	https://paperswithcode.com/paper/190600800
Repo
Framework

Robust parametric modeling of Alzheimer’s disease progression


Title	Robust parametric modeling of Alzheimer’s disease progression
Authors	Mostafa Mehdipour Ghazi, Mads Nielsen, Akshay Pai, Marc Modat, M. Jorge Cardoso, Sébastien Ourselin, Lauge Sørensen
Abstract	Quantitative characterization of disease progression using longitudinal data can provide long-term predictions for the pathological stages of individuals. This work studies robust modeling of Alzheimer’s disease progression using parametric methods. The proposed method linearly maps the individual’s chronological age to a disease progression score (DPS) and robustly fits a constrained generalized logistic function to the longitudinal dynamics of a biomarker as a function of the DPS using M-estimation. Robustness of the estimates is quantified using bootstrapping via Monte Carlo resampling, and the inflection points are used to temporally order the modeled biomarkers in the disease course. Moreover, kernel density estimation is applied to the obtained DPSs for clinical status prediction using a Bayesian classifier. Different M-estimators and logistic functions, including a new generalized type proposed in this study, called modified Stannard, are evaluated on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database for robust modeling of volumetric MRI and PET biomarkers, as well as cognitive tests. The results show that the modified Stannard function fitted using the modified Huber loss achieves the best modeling performance with an MAE of 0.071 across all biomarkers and bootstraps. In addition, applied to the ADNI test set, this model achieves a multi-class AUC of 0.87 in clinical status prediction, and it significantly outperforms an analogous state-of-the-art method with a biomarker modeling MAE of 0.071 vs. 0.073 (p < 0.001). Finally, the experiments show that the proposed model, trained using abundant ADNI data, generalizes well to data from the independent National Alzheimer’s Coordinating Center (NACC) database, where modeling performance is significantly improved (p < 0.001) compared with using a model trained on relatively scarce NACC data.
Tasks	Density Estimation
Published	2019-08-14
URL	https://arxiv.org/abs/1908.05338v2
PDF	https://arxiv.org/pdf/1908.05338v2.pdf
PWC	https://paperswithcode.com/paper/robust-parametric-modeling-of-alzheimers
Repo
Framework

Redundancy-Free Computation Graphs for Graph Neural Networks


Title	Redundancy-Free Computation Graphs for Graph Neural Networks
Authors	Zhihao Jia, Sina Lin, Rex Ying, Jiaxuan You, Jure Leskovec, Alex Aiken
Abstract	Graph Neural Networks (GNNs) are based on repeated aggregations of information across nodes’ neighbors in a graph. However, because common neighbors are shared between different nodes, this leads to repeated and inefficient computations. We propose Hierarchically Aggregated computation Graphs (HAGs), a new GNN graph representation that explicitly avoids redundancy by managing intermediate aggregation results hierarchically, eliminating repeated computations and unnecessary data transfers in GNN training and inference. We introduce an accurate cost function to quantitatively evaluate the runtime performance of different HAGs and use a novel HAG search algorithm to find optimized HAGs. Experiments show that the HAG representation significantly outperforms the standard GNN graph representation by increasing the end-to-end training throughput by up to 2.8x and reducing the aggregations and data transfers in GNN training by up to 6.3x and 5.6x, while maintaining the original model accuracy.
Tasks
Published	2019-06-09
URL	https://arxiv.org/abs/1906.03707v1
PDF	https://arxiv.org/pdf/1906.03707v1.pdf
PWC	https://paperswithcode.com/paper/redundancy-free-computation-graphs-for-graph
Repo
Framework

Modeling Cyber-Physical Human Systems via an Interplay Between Reinforcement Learning and Game Theory


Title	Modeling Cyber-Physical Human Systems via an Interplay Between Reinforcement Learning and Game Theory
Authors	Mert Albaba, Yildiray Yildiz
Abstract	Predicting the outcomes of cyber-physical systems with multiple human interactions is a challenging problem. This article reviews a game theoretical approach to address this issue, where reinforcement learning is employed to predict the time-extended interaction dynamics. We explain that the most attractive feature of the method is proposing a computationally feasible approach to simultaneously model multiple humans as decision makers, instead of determining the decision dynamics of the intelligent agent of interest and forcing the others to obey certain kinematic and dynamic constraints imposed by the environment. We present two recent exploitations of the method to model 1) unmanned aircraft integration into the National Airspace System and 2) highway traffic. We conclude the article by providing ongoing and future work about employing, improving and validating the method. We also provide related open problems and research opportunities.
Tasks
Published	2019-10-11
URL	https://arxiv.org/abs/1910.05092v1
PDF	https://arxiv.org/pdf/1910.05092v1.pdf
PWC	https://paperswithcode.com/paper/modeling-cyber-physical-human-systems-via-an
Repo
Framework

Steepest ascent can be exponential in bounded treewidth problems


Title	Steepest ascent can be exponential in bounded treewidth problems
Authors	David A. Cohen, Martin C. Cooper, Artem Kaznatcheev, Mark Wallace
Abstract	We investigate the complexity of local search based on steepest ascent. We show that even when all variables have domains of size two and the underlying constraint graph of variable interactions has bounded treewidth (in our construction, treewidth 7), there are fitness landscapes for which an exponential number of steps may be required to reach a local optimum. This is an improvement on prior recursive constructions of long steepest ascents, which we prove to need constraint graphs of unbounded treewidth.
Tasks
Published	2019-11-19
URL	https://arxiv.org/abs/1911.08600v2
PDF	https://arxiv.org/pdf/1911.08600v2.pdf
PWC	https://paperswithcode.com/paper/steepest-ascent-can-be-exponential-in-bounded
Repo
Framework

Simple Physical Adversarial Examples against End-to-End Autonomous Driving Models


Title	Simple Physical Adversarial Examples against End-to-End Autonomous Driving Models
Authors	Adith Boloor, Xin He, Christopher Gill, Yevgeniy Vorobeychik, Xuan Zhang
Abstract	Recent advances in machine learning, especially techniques such as deep neural networks, are promoting a range of high-stakes applications, including autonomous driving, which often relies on deep learning for perception. While deep learning for perception has been shown to be vulnerable to a host of subtle adversarial manipulations of images, end-to-end demonstrations of successful attacks, which manipulate the physical environment and result in physical consequences, are scarce. Moreover, attacks typically involve carefully constructed adversarial examples at the level of pixels. We demonstrate the first end-to-end attacks on autonomous driving in simulation, using simple physically realizable attacks: the painting of black lines on the road. These attacks target deep neural network models for end-to-end autonomous driving control. A systematic investigation shows that such attacks are surprisingly easy to engineer, and we describe scenarios (e.g., right turns) in which they are highly effective, and others that are less vulnerable (e.g., driving straight). Further, we use network deconvolution to demonstrate that the attacks succeed by inducing activation patterns similar to entirely different scenarios used in training.
Tasks	Autonomous Driving
Published	2019-03-12
URL	http://arxiv.org/abs/1903.05157v1
PDF	http://arxiv.org/pdf/1903.05157v1.pdf
PWC	https://paperswithcode.com/paper/simple-physical-adversarial-examples-against
Repo
Framework

Quantum Hamiltonian-Based Models and the Variational Quantum Thermalizer Algorithm


Title	Quantum Hamiltonian-Based Models and the Variational Quantum Thermalizer Algorithm
Authors	Guillaume Verdon, Jacob Marks, Sasha Nanda, Stefan Leichenauer, Jack Hidary
Abstract	We introduce a new class of generative quantum-neural-network-based models called Quantum Hamiltonian-Based Models (QHBMs). In doing so, we establish a paradigmatic approach for quantum-probabilistic hybrid variational learning, where we efficiently decompose the tasks of learning classical and quantum correlations in a way which maximizes the utility of both classical and quantum processors. In addition, we introduce the Variational Quantum Thermalizer (VQT) for generating the thermal state of a given Hamiltonian and target temperature, a task for which QHBMs are naturally well-suited. The VQT can be seen as a generalization of the Variational Quantum Eigensolver (VQE) to thermal states: we show that the VQT converges to the VQE in the zero temperature limit. We provide numerical results demonstrating the efficacy of these techniques in illustrative examples. We use QHBMs and the VQT on Heisenberg spin systems, we apply QHBMs to learn entanglement Hamiltonians and compression codes in simulated free Bosonic systems, and finally we use the VQT to prepare thermal Fermionic Gaussian states for quantum simulation.
Tasks
Published	2019-10-04
URL	https://arxiv.org/abs/1910.02071v1
PDF	https://arxiv.org/pdf/1910.02071v1.pdf
PWC	https://paperswithcode.com/paper/quantum-hamiltonian-based-models-and-the
Repo
Framework

Associating Natural Language Comment and Source Code Entities


Title	Associating Natural Language Comment and Source Code Entities
Authors	Sheena Panthaplackel, Milos Gligoric, Raymond J. Mooney, Junyi Jessy Li
Abstract	Comments are an integral part of software development; they are natural language descriptions associated with source code elements. Understanding explicit associations can be useful in improving code comprehensibility and maintaining the consistency between code and comments. As an initial step towards this larger goal, we address the task of associating entities in Javadoc comments with elements in Java source code. We propose an approach for automatically extracting supervised data using revision histories of open source projects and present a manually annotated evaluation dataset for this task. We develop a binary classifier and a sequence labeling model by crafting a rich feature set which encompasses various aspects of code, comments, and the relationships between them. Experiments show that our systems outperform several baselines learning from the proposed supervision.
Tasks
Published	2019-12-13
URL	https://arxiv.org/abs/1912.06728v1
PDF	https://arxiv.org/pdf/1912.06728v1.pdf
PWC	https://paperswithcode.com/paper/associating-natural-language-comment-and
Repo
Framework

DeepFuse: An IMU-Aware Network for Real-Time 3D Human Pose Estimation from Multi-View Image


Title	DeepFuse: An IMU-Aware Network for Real-Time 3D Human Pose Estimation from Multi-View Image
Authors	Fuyang Huang, Ailing Zeng, Minhao Liu, Qiuxia Lai, Qiang Xu
Abstract	In this paper, we propose a two-stage fully 3D network, namely \textbf{DeepFuse}, to estimate human pose in 3D space by fusing body-worn Inertial Measurement Unit (IMU) data and multi-view images deeply. The first stage is designed for pure vision estimation. To preserve data primitiveness of multi-view inputs, the vision stage uses multi-channel volume as data representation and 3D soft-argmax as activation layer. The second one is the IMU refinement stage which introduces an IMU-bone layer to fuse the IMU and vision data earlier at data level. without requiring a given skeleton model a priori, we can achieve a mean joint error of $28.9$mm on TotalCapture dataset and $13.4$mm on Human3.6M dataset under protocol 1, improving the SOTA result by a large margin. Finally, we discuss the effectiveness of a fully 3D network for 3D pose estimation experimentally which may benefit future research.
Tasks	3D Human Pose Estimation, 3D Pose Estimation, Pose Estimation
Published	2019-12-09
URL	https://arxiv.org/abs/1912.04071v1
PDF	https://arxiv.org/pdf/1912.04071v1.pdf
PWC	https://paperswithcode.com/paper/deepfuse-an-imu-aware-network-for-real-time
Repo
Framework