July 28, 2019

3127 words 15 mins read

Paper Group ANR 426

Anomaly Detection in a Digital Video Broadcasting System Using Timed Automata. Pseudo-positive regularization for deep person re-identification. Hierarchical Implicit Models and Likelihood-Free Variational Inference. Machine Learning Approach for Detection of nonTor Traffic. Learning Disentangling and Fusing Networks for Face Completion Under Struc …

Anomaly Detection in a Digital Video Broadcasting System Using Timed Automata


Title	Anomaly Detection in a Digital Video Broadcasting System Using Timed Automata
Authors	Xiaoran Liu, Qin Lin, Sicco Verwer, Dmitri Jarnikov
Abstract	This paper focuses on detecting anomalies in a digital video broadcasting (DVB) system from providers’ perspective. We learn a probabilistic deterministic real timed automaton profiling benign behavior of encryption control in the DVB control access system. This profile is used as a one-class classifier. Anomalous items in a testing sequence are detected when the sequence is not accepted by the learned model.
Tasks	Anomaly Detection, One-class classifier
Published	2017-05-24
URL	http://arxiv.org/abs/1705.09650v1
PDF	http://arxiv.org/pdf/1705.09650v1.pdf
PWC	https://paperswithcode.com/paper/anomaly-detection-in-a-digital-video
Repo
Framework

Pseudo-positive regularization for deep person re-identification


Title	Pseudo-positive regularization for deep person re-identification
Authors	Fuqing Zhu, Xiangwei Kong, Haiyan Fu, Qi Tian
Abstract	An intrinsic challenge of person re-identification (re-ID) is the annotation difficulty. This typically means 1) few training samples per identity, and 2) thus the lack of diversity among the training samples. Consequently, we face high risk of over-fitting when training the convolutional neural network (CNN), a state-of-the-art method in person re-ID. To reduce the risk of over-fitting, this paper proposes a Pseudo Positive Regularization (PPR) method to enrich the diversity of the training data. Specifically, unlabeled data from an independent pedestrian database is retrieved using the target training data as query. A small proportion of these retrieved samples are randomly selected as the Pseudo Positive samples and added to the target training set for the supervised CNN training. The addition of Pseudo Positive samples is therefore a data augmentation method to reduce the risk of over-fitting during CNN training. We implement our idea in the identification CNN models (i.e., CaffeNet, VGGNet-16 and ResNet-50). On CUHK03 and Market-1501 datasets, experimental results demonstrate that the proposed method consistently improves the baseline and yields competitive performance to the state-of-the-art person re-ID methods.
Tasks	Data Augmentation, Person Re-Identification
Published	2017-11-17
URL	http://arxiv.org/abs/1711.06500v1
PDF	http://arxiv.org/pdf/1711.06500v1.pdf
PWC	https://paperswithcode.com/paper/pseudo-positive-regularization-for-deep
Repo
Framework

Hierarchical Implicit Models and Likelihood-Free Variational Inference


Title	Hierarchical Implicit Models and Likelihood-Free Variational Inference
Authors	Dustin Tran, Rajesh Ranganath, David M. Blei
Abstract	Implicit probabilistic models are a flexible class of models defined by a simulation process for data. They form the basis for theories which encompass our understanding of the physical world. Despite this fundamental nature, the use of implicit models remains limited due to challenges in specifying complex latent structure in them, and in performing inferences in such models with large data sets. In this paper, we first introduce hierarchical implicit models (HIMs). HIMs combine the idea of implicit densities with hierarchical Bayesian modeling, thereby defining models via simulators of data with rich hidden structure. Next, we develop likelihood-free variational inference (LFVI), a scalable variational inference algorithm for HIMs. Key to LFVI is specifying a variational family that is also implicit. This matches the model’s flexibility and allows for accurate approximation of the posterior. We demonstrate diverse applications: a large-scale physical simulator for predator-prey populations in ecology; a Bayesian generative adversarial network for discrete data; and a deep implicit model for text generation.
Tasks	Text Generation
Published	2017-02-28
URL	http://arxiv.org/abs/1702.08896v3
PDF	http://arxiv.org/pdf/1702.08896v3.pdf
PWC	https://paperswithcode.com/paper/hierarchical-implicit-models-and-likelihood
Repo
Framework

Machine Learning Approach for Detection of nonTor Traffic


Title	Machine Learning Approach for Detection of nonTor Traffic
Authors	Elike Hodo, Xavier Bellekens, Ephraim Iorkyase, Andrew Hamilton, Christos Tachtatzis, Robert Atkinson
Abstract	Intrusion detection has attracted a considerable interest from researchers and industries. After many years of research the community still faces the problem of building reliable and efficient intrusion detection systems (IDS) capable of handling large quantities of data with changing patterns in real time situations. The Tor network is popular in providing privacy and security to end user by anonymising the identity of internet users connecting through a series of tunnels and nodes. This work focuses on the classification of Tor traffic and nonTor traffic to expose the activities within Tor traffic that minimizes the protection of users. A study to compare the reliability and efficiency of Artificial Neural Network and Support vector machine in detecting nonTor traffic in UNB-CIC Tor Network Traffic dataset is presented in this paper. The results are analysed based on the overall accuracy, detection rate and false positive rate of the two algorithms. Experimental results show that both algorithms could detect nonTor traffic in the dataset. A hybrid Artificial neural network proved a better classifier than SVM in detecting nonTor traffic in UNB-CIC Tor Network Traffic dataset.
Tasks	Intrusion Detection
Published	2017-08-29
URL	http://arxiv.org/abs/1708.08725v1
PDF	http://arxiv.org/pdf/1708.08725v1.pdf
PWC	https://paperswithcode.com/paper/machine-learning-approach-for-detection-of
Repo
Framework

Learning Disentangling and Fusing Networks for Face Completion Under Structured Occlusions


Title	Learning Disentangling and Fusing Networks for Face Completion Under Structured Occlusions
Authors	Zhihang Li, Yibo Hu, Ran He
Abstract	Face completion aims to generate semantically new pixels for missing facial components. It is a challenging generative task due to large variations of face appearance. This paper studies generative face completion under structured occlusions. We treat the face completion and corruption as disentangling and fusing processes of clean faces and occlusions, and propose a jointly disentangling and fusing Generative Adversarial Network (DF-GAN). First, three domains are constructed, corresponding to the distributions of occluded faces, clean faces and structured occlusions. The disentangling and fusing processes are formulated as the transformations between the three domains. Then the disentangling and fusing networks are built to learn the transformations from unpaired data, where the encoder-decoder structure is adopted and allows DF-GAN to simulate structure occlusions by modifying the latent representations. Finally, the disentangling and fusing processes are unified into a dual learning framework along with an adversarial strategy. The proposed method is evaluated on Meshface verification problem. Experimental results on four Meshface databases demonstrate the effectiveness of our proposed method for the face completion under structured occlusions.
Tasks	Facial Inpainting
Published	2017-12-13
URL	http://arxiv.org/abs/1712.04646v1
PDF	http://arxiv.org/pdf/1712.04646v1.pdf
PWC	https://paperswithcode.com/paper/learning-disentangling-and-fusing-networks
Repo
Framework

Learning to Plan Chemical Syntheses


Title	Learning to Plan Chemical Syntheses
Authors	Marwin H. S. Segler, Mike Preuss, Mark P. Waller
Abstract	From medicines to materials, small organic molecules are indispensable for human well-being. To plan their syntheses, chemists employ a problem solving technique called retrosynthesis. In retrosynthesis, target molecules are recursively transformed into increasingly simpler precursor compounds until a set of readily available starting materials is obtained. Computer-aided retrosynthesis would be a highly valuable tool, however, past approaches were slow and provided results of unsatisfactory quality. Here, we employ Monte Carlo Tree Search (MCTS) to efficiently discover retrosynthetic routes. MCTS was combined with an expansion policy network that guides the search, and an “in-scope” filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on 12 million reactions, which represents essentially all reactions ever published in organic chemistry. Our system solves almost twice as many molecules and is 30 times faster in comparison to the traditional search method based on extracted rules and hand-coded heuristics. Finally after a 60 year history of computer-aided synthesis planning, chemists can no longer distinguish between routes generated by a computer system and real routes taken from the scientific literature. We anticipate that our method will accelerate drug and materials discovery by assisting chemists to plan better syntheses faster, and by enabling fully automated robot synthesis.
Tasks
Published	2017-08-14
URL	http://arxiv.org/abs/1708.04202v1
PDF	http://arxiv.org/pdf/1708.04202v1.pdf
PWC	https://paperswithcode.com/paper/learning-to-plan-chemical-syntheses
Repo
Framework

Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Training


Title	Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Training
Authors	Faisal Mahmood, Richard Chen, Nicholas J. Durr
Abstract	To realize the full potential of deep learning for medical imaging, large annotated datasets are required for training. Such datasets are difficult to acquire because labeled medical images are not usually available due to privacy issues, lack of experts available for annotation, underrepresentation of rare conditions and poor standardization. Lack of annotated data has been addressed in conventional vision applications using synthetic images refined via unsupervised adversarial training to look like real images. However, this approach is difficult to extend to general medical imaging because of the complex and diverse set of features found in real human tissues. We propose an alternative framework that uses a reverse flow, where adversarial training is used to make real medical images more like synthetic images, and hypothesize that clinically-relevant features can be preserved via self-regularization. These domain-adapted images can then be accurately interpreted by networks trained on large datasets of synthetic medical images. We test this approach for the notoriously difficult task of depth-estimation from endoscopy. We train a depth estimator on a large dataset of synthetic images generated using an accurate forward model of an endoscope and an anatomically-realistic colon. This network predicts significantly better depths when using synthetic-like domain-adapted images compared to the real images, confirming that the clinically-relevant features of depth are preserved.
Tasks	Depth Estimation, Domain Adaptation
Published	2017-11-17
URL	http://arxiv.org/abs/1711.06606v2
PDF	http://arxiv.org/pdf/1711.06606v2.pdf
PWC	https://paperswithcode.com/paper/unsupervised-reverse-domain-adaptation-for
Repo
Framework

CNN-MERP: An FPGA-Based Memory-Efficient Reconfigurable Processor for Forward and Backward Propagation of Convolutional Neural Networks


Title	CNN-MERP: An FPGA-Based Memory-Efficient Reconfigurable Processor for Forward and Backward Propagation of Convolutional Neural Networks
Authors	Xushen Han, Dajiang Zhou, Shihao Wang, Shinji Kimura
Abstract	Large-scale deep convolutional neural networks (CNNs) are widely used in machine learning applications. While CNNs involve huge complexity, VLSI (ASIC and FPGA) chips that deliver high-density integration of computational resources are regarded as a promising platform for CNN’s implementation. At massive parallelism of computational units, however, the external memory bandwidth, which is constrained by the pin count of the VLSI chip, becomes the system bottleneck. Moreover, VLSI solutions are usually regarded as a lack of the flexibility to be reconfigured for the various parameters of CNNs. This paper presents CNN-MERP to address these issues. CNN-MERP incorporates an efficient memory hierarchy that significantly reduces the bandwidth requirements from multiple optimizations including on/off-chip data allocation, data flow optimization and data reuse. The proposed 2-level reconfigurability is utilized to enable fast and efficient reconfiguration, which is based on the control logic and the multiboot feature of FPGA. As a result, an external memory bandwidth requirement of 1.94MB/GFlop is achieved, which is 55% lower than prior arts. Under limited DRAM bandwidth, a system throughput of 1244GFlop/s is achieved at the Vertex UltraScale platform, which is 5.48 times higher than the state-of-the-art FPGA implementations.
Tasks
Published	2017-03-22
URL	http://arxiv.org/abs/1703.07348v1
PDF	http://arxiv.org/pdf/1703.07348v1.pdf
PWC	https://paperswithcode.com/paper/cnn-merp-an-fpga-based-memory-efficient
Repo
Framework

Latent Geometry and Memorization in Generative Models


Title	Latent Geometry and Memorization in Generative Models
Authors	Matt Feiszli
Abstract	It can be difficult to tell whether a trained generative model has learned to generate novel examples or has simply memorized a specific set of outputs. In published work, it is common to attempt to address this visually, for example by displaying a generated example and its nearest neighbor(s) in the training set (in, for example, the L2 metric). As any generative model induces a probability density on its output domain, we propose studying this density directly. We first study the geometry of the latent representation and generator, relate this to the output density, and then develop techniques to compute and inspect the output density. As an application, we demonstrate that “memorization” tends to a density made of delta functions concentrated on the memorized examples. We note that without first understanding the geometry, the measurement would be essentially impossible to make.
Tasks
Published	2017-05-25
URL	http://arxiv.org/abs/1705.09303v1
PDF	http://arxiv.org/pdf/1705.09303v1.pdf
PWC	https://paperswithcode.com/paper/latent-geometry-and-memorization-in
Repo
Framework

Unsupervised Person Re-identification: Clustering and Fine-tuning


Title	Unsupervised Person Re-identification: Clustering and Fine-tuning
Authors	Hehe Fan, Liang Zheng, Yi Yang
Abstract	The superiority of deeply learned pedestrian representations has been reported in very recent literature of person re-identification (re-ID). In this paper, we consider the more pragmatic issue of learning a deep feature with no or only a few labels. We propose a progressive unsupervised learning (PUL) method to transfer pretrained deep representations to unseen domains. Our method is easy to implement and can be viewed as an effective baseline for unsupervised re-ID feature learning. Specifically, PUL iterates between 1) pedestrian clustering and 2) fine-tuning of the convolutional neural network (CNN) to improve the original model trained on the irrelevant labeled dataset. Since the clustering results can be very noisy, we add a selection operation between the clustering and fine-tuning. At the beginning when the model is weak, CNN is fine-tuned on a small amount of reliable examples which locate near to cluster centroids in the feature space. As the model becomes stronger in subsequent iterations, more images are being adaptively selected as CNN training samples. Progressively, pedestrian clustering and the CNN model are improved simultaneously until algorithm convergence. This process is naturally formulated as self-paced learning. We then point out promising directions that may lead to further improvement. Extensive experiments on three large-scale re-ID datasets demonstrate that PUL outputs discriminative features that improve the re-ID accuracy.
Tasks	Person Re-Identification, Unsupervised Person Re-Identification
Published	2017-05-30
URL	http://arxiv.org/abs/1705.10444v2
PDF	http://arxiv.org/pdf/1705.10444v2.pdf
PWC	https://paperswithcode.com/paper/unsupervised-person-re-identification
Repo
Framework

Cooperative Learning with Visual Attributes


Title	Cooperative Learning with Visual Attributes
Authors	Tanmay Batra, Devi Parikh
Abstract	Learning paradigms involving varying levels of supervision have received a lot of interest within the computer vision and machine learning communities. The supervisory information is typically considered to come from a human supervisor – a “teacher” figure. In this paper, we consider an alternate source of supervision – a “peer” – i.e. a different machine. We introduce cooperative learning, where two agents trying to learn the same visual concepts, but in potentially different environments using different sources of data (sensors), communicate their current knowledge of these concepts to each other. Given the distinct sources of data in both agents, the mode of communication between the two agents is not obvious. We propose the use of visual attributes – semantic mid-level visual properties such as furry, wooden, etc.– as the mode of communication between the agents. Our experiments in three domains – objects, scenes, and animals – demonstrate that our proposed cooperative learning approach improves the performance of both agents as compared to their performance if they were to learn in isolation. Our approach is particularly applicable in scenarios where privacy, security and/or bandwidth constraints restrict the amount and type of information the two agents can exchange.
Tasks
Published	2017-05-16
URL	http://arxiv.org/abs/1705.05512v1
PDF	http://arxiv.org/pdf/1705.05512v1.pdf
PWC	https://paperswithcode.com/paper/cooperative-learning-with-visual-attributes
Repo
Framework


Title	Beautiful and damned. Combined effect of content quality and social ties on user engagement
Authors	Luca M. Aiello, Rossano Schifanella, Miriam Redi, Stacey Svetlichnaya, Frank Liu, Simon Osindero
Abstract	User participation in online communities is driven by the intertwinement of the social network structure with the crowd-generated content that flows along its links. These aspects are rarely explored jointly and at scale. By looking at how users generate and access pictures of varying beauty on Flickr, we investigate how the production of quality impacts the dynamics of online social systems. We develop a deep learning computer vision model to score images according to their aesthetic value and we validate its output through crowdsourcing. By applying it to over 15B Flickr photos, we study for the first time how image beauty is distributed over a large-scale social system. Beautiful images are evenly distributed in the network, although only a small core of people get social recognition for them. To study the impact of exposure to quality on user engagement, we set up matching experiments aimed at detecting causality from observational data. Exposure to beauty is double-edged: following people who produce high-quality content increases one’s probability of uploading better photos; however, an excessive imbalance between the quality generated by a user and the user’s neighbors leads to a decline in engagement. Our analysis has practical implications for improving link recommender systems.
Tasks	Recommendation Systems
Published	2017-11-01
URL	http://arxiv.org/abs/1711.00536v1
PDF	http://arxiv.org/pdf/1711.00536v1.pdf
PWC	https://paperswithcode.com/paper/beautiful-and-damned-combined-effect-of
Repo
Framework

Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators


Title	Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators
Authors	Mario Lezcano Casado, Atilim Gunes Baydin, David Martinez Rubio, Tuan Anh Le, Frank Wood, Lukas Heinrich, Gilles Louppe, Kyle Cranmer, Karen Ng, Wahid Bhimji, Prabhat
Abstract	We consider the problem of Bayesian inference in the family of probabilistic models implicitly defined by stochastic generative models of data. In scientific fields ranging from population biology to cosmology, low-level mechanistic components are composed to create complex generative models. These models lead to intractable likelihoods and are typically non-differentiable, which poses challenges for traditional approaches to inference. We extend previous work in “inference compilation”, which combines universal probabilistic programming and deep learning methods, to large-scale scientific simulators, and introduce a C++ based probabilistic programming library called CPProb. We successfully use CPProb to interface with SHERPA, a large code-base used in particle physics. Here we describe the technical innovations realized and planned for this library.
Tasks	Bayesian Inference, Probabilistic Programming
Published	2017-12-21
URL	http://arxiv.org/abs/1712.07901v1
PDF	http://arxiv.org/pdf/1712.07901v1.pdf
PWC	https://paperswithcode.com/paper/improvements-to-inference-compilation-for
Repo
Framework

Temporal Modeling Approaches for Large-scale Youtube-8M Video Understanding


Title	Temporal Modeling Approaches for Large-scale Youtube-8M Video Understanding
Authors	Fu Li, Chuang Gan, Xiao Liu, Yunlong Bian, Xiang Long, Yandong Li, Zhichao Li, Jie Zhou, Shilei Wen
Abstract	This paper describes our solution for the video recognition task of the Google Cloud and YouTube-8M Video Understanding Challenge that ranked the 3rd place. Because the challenge provides pre-extracted visual and audio features instead of the raw videos, we mainly investigate various temporal modeling approaches to aggregate the frame-level features for multi-label video recognition. Our system contains three major components: two-stream sequence model, fast-forward sequence model and temporal residual neural networks. Experiment results on the challenging Youtube-8M dataset demonstrate that our proposed temporal modeling approaches can significantly improve existing temporal modeling approaches in the large-scale video recognition tasks. To be noted, our fast-forward LSTM with a depth of 7 layers achieves 82.75% in term of GAP@20 on the Kaggle Public test set.
Tasks	Video Recognition, Video Understanding
Published	2017-07-14
URL	http://arxiv.org/abs/1707.04555v1
PDF	http://arxiv.org/pdf/1707.04555v1.pdf
PWC	https://paperswithcode.com/paper/temporal-modeling-approaches-for-large-scale
Repo
Framework

An MPI-Based Python Framework for Distributed Training with Keras


Title	An MPI-Based Python Framework for Distributed Training with Keras
Authors	Dustin Anderson, Jean-Roch Vlimant, Maria Spiropulu
Abstract	We present a lightweight Python framework for distributed training of neural networks on multiple GPUs or CPUs. The framework is built on the popular Keras machine learning library. The Message Passing Interface (MPI) protocol is used to coordinate the training process, and the system is well suited for job submission at supercomputing sites. We detail the software’s features, describe its use, and demonstrate its performance on systems of varying sizes on a benchmark problem drawn from high-energy physics research.
Tasks
Published	2017-12-16
URL	http://arxiv.org/abs/1712.05878v1
PDF	http://arxiv.org/pdf/1712.05878v1.pdf
PWC	https://paperswithcode.com/paper/an-mpi-based-python-framework-for-distributed
Repo
Framework