January 30, 2020

3295 words 16 mins read

Paper Group ANR 308

Model Bridging: To Interpretable Simulation Model From Neural Network. Generalised Zero-Shot Learning with a Classifier Ensemble over Multi-Modal Embedding Spaces. PaintBot: A Reinforcement Learning Approach for Natural Media Painting. Where Are We? Using Scopus to Map the Literature at the Intersection Between Artificial Intelligence and Crime. Mu …

Model Bridging: To Interpretable Simulation Model From Neural Network


Title	Model Bridging: To Interpretable Simulation Model From Neural Network
Authors	Keiichi Kisamori, Keisuke Yamazaki, Yuto Komori, Hiroshi Tokieda
Abstract	The interpretability of machine learning, particularly for deep neural networks, is strongly required when performing decision-making in a real-world application. There are several studies that show that interpretability is obtained by replacing a non-explainable neural network with an explainable simplified surrogate model. Meanwhile, another approach to understanding the target system is simulation modeled by human knowledge with interpretable simulation parameters. Recently developed simulation learning based on applications of kernel mean embedding is a method used to estimate simulation parameters as posterior distributions. However, there was no relation between the machine learning model and the simulation model. Furthermore, the computational cost of simulation learning is very expensive because of the complexity of the simulation model. To address these difficulties, we propose a “model bridging” framework to bridge machine learning models with simulation models by a series of kernel mean embeddings. The proposed framework enables us to obtain predictions and interpretable simulation parameters simultaneously without the computationally expensive calculations associated with simulations. In this study, we investigate a Bayesian neural network model with a few hidden layers serving as an un-explainable machine learning model. We apply the proposed framework to production simulation and simulation of fluid-dynamics, which are important in the manufacturing industry.
Tasks	Decision Making
Published	2019-06-22
URL	https://arxiv.org/abs/1906.09391v2
PDF	https://arxiv.org/pdf/1906.09391v2.pdf
PWC	https://paperswithcode.com/paper/model-bridging-to-interpretable-simulation
Repo
Framework


Title	Generalised Zero-Shot Learning with a Classifier Ensemble over Multi-Modal Embedding Spaces
Authors	Rafael Felix, Ben Harwood, Michele Sasdelli, Gustavo Carneiro
Abstract	Generalised zero-shot learning (GZSL) methods aim to classify previously seen and unseen visual classes by leveraging the semantic information of those classes. In the context of GZSL, semantic information is non-visual data such as a text description of both seen and unseen classes. Previous GZSL methods have utilised transformations between visual and semantic embedding spaces, as well as the learning of joint spaces that include both visual and semantic information. In either case, classification is then performed on a single learned space. We argue that each embedding space contains complementary information for the GZSL problem. By using just a visual, semantic or joint space some of this information will invariably be lost. In this paper, we demonstrate the advantages of our new GZSL method that combines the classification of visual, semantic and joint spaces. Most importantly, this ensembling allows for more information from the source domains to be seen during classification. An additional contribution of our work is the application of a calibration procedure for each classifier in the ensemble. This calibration mitigates the problem of model selection when combining the classifiers. Lastly, our proposed method achieves state-of-the-art results on the CUB, AWA1 and AWA2 benchmark data sets and provides competitive performance on the SUN data set.
Tasks	Calibration, Model Selection, Zero-Shot Learning
Published	2019-08-06
URL	https://arxiv.org/abs/1908.02013v1
PDF	https://arxiv.org/pdf/1908.02013v1.pdf
PWC	https://paperswithcode.com/paper/generalised-zero-shot-learning-with-a
Repo
Framework

PaintBot: A Reinforcement Learning Approach for Natural Media Painting


Title	PaintBot: A Reinforcement Learning Approach for Natural Media Painting
Authors	Biao Jia, Chen Fang, Jonathan Brandt, Byungmoon Kim, Dinesh Manocha
Abstract	We propose a new automated digital painting framework, based on a painting agent trained through reinforcement learning. To synthesize an image, the agent selects a sequence of continuous-valued actions representing primitive painting strokes, which are accumulated on a digital canvas. Action selection is guided by a given reference image, which the agent attempts to replicate subject to the limitations of the action space and the agent’s learned policy. The painting agent policy is determined using a variant of proximal policy optimization reinforcement learning. During training, our agent is presented with patches sampled from an ensemble of reference images. To accelerate training convergence, we adopt a curriculum learning strategy, whereby reference patches are sampled according to how challenging they are using the current policy. We experiment with differing loss functions, including pixel-wise and perceptual loss, which have consequent differing effects on the learned policy. We demonstrate that our painting agent can learn an effective policy with a high dimensional continuous action space comprising pen pressure, width, tilt, and color, for a variety of painting styles. Through a coarse-to-fine refinement process our agent can paint arbitrarily complex images in the desired style.
Tasks
Published	2019-04-03
URL	http://arxiv.org/abs/1904.02201v1
PDF	http://arxiv.org/pdf/1904.02201v1.pdf
PWC	https://paperswithcode.com/paper/paintbot-a-reinforcement-learning-approach
Repo
Framework

Where Are We? Using Scopus to Map the Literature at the Intersection Between Artificial Intelligence and Crime


Title	Where Are We? Using Scopus to Map the Literature at the Intersection Between Artificial Intelligence and Crime
Authors	Gian Maria Campedelli
Abstract	Research on Artificial Intelligence (AI) applications has spread over many scientific disciplines. Scientists have tested the power of intelligent algorithms developed to predict (or learn from) natural, physical and social phenomena. This also applies to crime-related research problems. Nonetheless, studies that map the current state of the art at the intersection between AI and crime are lacking. What are the current research trends in terms of topics in this area? What is the structure of scientific collaboration when considering works investigating criminal issues using machine learning, deep learning and AI in general? What are the most active countries in this specific scientific sphere? Using data retrieved from Scopus database, this work quantitatively analyzes published works at the intersection between AI and crime employing network science to respond to these questions. Results show that researchers are mainly focusing on cyber-related criminal topics and that relevant themes such as algorithmic discrimination, fairness, and ethics are considerably overlooked. Furthermore, data highlight the extremely disconnected structure of co-authorship networks. Such disconnectedness may represent a substantial obstacle to a more solid community of scientists interested in these topics. Additionally, the graph of scientific collaboration indicates that countries that are more prone to engage in international partnerships are generally less central in the network. This means that scholars working in highly productive countries (e.g. the United States, China) tend to collaborate with researchers based in their same countries. Finally, current issues and future developments within this scientific area are also discussed.
Tasks
Published	2019-12-23
URL	https://arxiv.org/abs/1912.11084v1
PDF	https://arxiv.org/pdf/1912.11084v1.pdf
PWC	https://paperswithcode.com/paper/where-are-we-using-scopus-to-map-the
Repo
Framework

Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation


Title	Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation
Authors	Risto Vuorio, Shao-Hua Sun, Hexiang Hu, Joseph J. Lim
Abstract	Model-agnostic meta-learners aim to acquire meta-learned parameters from similar tasks to adapt to novel tasks from the same distribution with few gradient updates. With the flexibility in the choice of models, those frameworks demonstrate appealing performance on a variety of domains such as few-shot image classification and reinforcement learning. However, one important limitation of such frameworks is that they seek a common initialization shared across the entire task distribution, substantially limiting the diversity of the task distributions that they are able to learn from. In this paper, we augment MAML with the capability to identify the mode of tasks sampled from a multimodal task distribution and adapt quickly through gradient updates. Specifically, we propose a multimodal MAML (MMAML) framework, which is able to modulate its meta-learned prior parameters according to the identified mode, allowing more efficient fast adaptation. We evaluate the proposed model on a diverse set of few-shot learning tasks, including regression, image classification, and reinforcement learning. The results not only demonstrate the effectiveness of our model in modulating the meta-learned prior in response to the characteristics of tasks but also show that training on a multimodal distribution can produce an improvement over unimodal training.
Tasks	Few-Shot Image Classification, Few-Shot Learning, Image Classification, Meta-Learning
Published	2019-10-30
URL	https://arxiv.org/abs/1910.13616v1
PDF	https://arxiv.org/pdf/1910.13616v1.pdf
PWC	https://paperswithcode.com/paper/multimodal-model-agnostic-meta-learning-via
Repo
Framework

Bridging the Knowledge Gap: Enhancing Question Answering with World and Domain Knowledge


Title	Bridging the Knowledge Gap: Enhancing Question Answering with World and Domain Knowledge
Authors	Travis R. Goodwin, Dina Demner-Fushman
Abstract	In this paper we present OSCAR (Ontology-based Semantic Composition Augmented Regularization), a method for injecting task-agnostic knowledge from an Ontology or knowledge graph into a neural network during pretraining. We evaluated the impact of including OSCAR when pretraining BERT with Wikipedia articles by measuring the performance when fine-tuning on two question answering tasks involving world knowledge and causal reasoning and one requiring domain (healthcare) knowledge and obtained 33:3%, 18:6%, and 4% improved accuracy compared to pretraining BERT without OSCAR and obtaining new state-of-the-art results on two of the tasks.
Tasks	Question Answering, Semantic Composition
Published	2019-10-16
URL	https://arxiv.org/abs/1910.07429v1
PDF	https://arxiv.org/pdf/1910.07429v1.pdf
PWC	https://paperswithcode.com/paper/bridging-the-knowledge-gap-enhancing-question
Repo
Framework

Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attacks


Title	Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attacks
Authors	Jiadong Lin, Chuanbiao Song, Kun He, Liwei Wang, John E. Hopcroft
Abstract	Deep learning models are vulnerable to adversarial examples crafted by applying human-imperceptible perturbations on benign inputs. However, under the black-box setting, most existing adversaries often have a poor transferability to attack other defense models. In this work, from the perspective of regarding the adversarial example generation as an optimization process, we propose two new methods to improve the transferability of adversarial examples, namely Nesterov Iterative Fast Gradient Sign Method (NI-FGSM) and Scale-Invariant attack Method (SIM). NI-FGSM aims to adapt Nesterov accelerated gradient into the iterative attacks so as to effectively look ahead and improve the transferability of adversarial examples. While SIM is based on our discovery on the scale-invariant property of deep learning models, for which we leverage to optimize the adversarial perturbations over the scale copies of the input images so as to avoid “overfitting” on the white-box model being attacked and generate more transferable adversarial examples. NI-FGSM and SIM can be naturally integrated to build a robust gradient-based attack to generate more transferable adversarial examples against the defense models. Empirical results on ImageNet dataset demonstrate that our attack methods exhibit higher transferability and achieve higher attack success rates than state-of-the-art gradient-based attacks.
Tasks	Adversarial Attack
Published	2019-08-17
URL	https://arxiv.org/abs/1908.06281v5
PDF	https://arxiv.org/pdf/1908.06281v5.pdf
PWC	https://paperswithcode.com/paper/nesterov-accelerated-gradient-and-scale
Repo
Framework

Directional Regularized Tensor Modeling for Video Rain Streaks Removal


Title	Directional Regularized Tensor Modeling for Video Rain Streaks Removal
Authors	Zhaoyang Sun, Shengwu Xiong, Ryan Wen Liu
Abstract	Outdoor videos sometimes contain unexpected rain streaks due to the rainy weather, which bring negative effects on subsequent computer vision applications, e.g., video surveillance, object recognition and tracking, etc. In this paper, we propose a directional regularized tensor-based video deraining model by taking into consideration the arbitrary direction of rain streaks. In particular, the sparsity of rain streaks in spatial and derivative domains, the spatiotemporal sparsity and low-rank property of video background are incorporated into the proposed method. Different from many previous methods under the assumption of vertically falling rain streaks, we consider a more realistic assumption that all the rain streaks in a video fall in an approximately similar arbitrary direction. The resulting complicated optimization problem will be effectively solved through an alternating direction method. Comprehensive experiments on both synthetic and realistic datasets have demonstrated the superiority of the proposed deraining method.
Tasks	Object Recognition, Rain Removal
Published	2019-02-19
URL	http://arxiv.org/abs/1902.07090v1
PDF	http://arxiv.org/pdf/1902.07090v1.pdf
PWC	https://paperswithcode.com/paper/directional-regularized-tensor-modeling-for
Repo
Framework

Encoding CT Anatomy Knowledge for Unpaired Chest X-ray Image Decomposition


Title	Encoding CT Anatomy Knowledge for Unpaired Chest X-ray Image Decomposition
Authors	Zeju Li, Han Li, Hu Han, Gonglei Shi, Jiannan Wang, S. Kevin Zhou
Abstract	Although chest X-ray (CXR) offers a 2D projection with overlapped anatomies, it is widely used for clinical diagnosis. There is clinical evidence supporting that decomposing an X-ray image into different components (e.g., bone, lung and soft tissue) improves diagnostic value. We hereby propose a decomposition generative adversarial network (DecGAN) to anatomically decompose a CXR image but with unpaired data. We leverage the anatomy knowledge embedded in CT, which features a 3D volume with clearly visible anatomies. Our key idea is to embed CT priori decomposition knowledge into the latent space of unpaired CXR autoencoder. Specifically, we train DecGAN with a decomposition loss, adversarial losses, cycle-consistency losses and a mask loss to guarantee that the decomposed results of the latent space preserve realistic body structures. Extensive experiments demonstrate that DecGAN provides superior unsupervised CXR bone suppression results and the feasibility of modulating CXR components by latent space disentanglement. Furthermore, we illustrate the diagnostic value of DecGAN and demonstrate that it outperforms the state-of-the-art approaches in terms of predicting 11 out of 14 common lung diseases.
Tasks
Published	2019-09-16
URL	https://arxiv.org/abs/1909.12922v1
PDF	https://arxiv.org/pdf/1909.12922v1.pdf
PWC	https://paperswithcode.com/paper/encoding-ct-anatomy-knowledge-for-unpaired
Repo
Framework

Automated Blood Cell Detection and Counting via Deep Learning for Microfluidic Point-of-Care Medical Devices


Title	Automated Blood Cell Detection and Counting via Deep Learning for Microfluidic Point-of-Care Medical Devices
Authors	Tiancheng Xia, Richard Jiang, YongQing Fu, Nanlin Jin
Abstract	Automated in-vitro cell detection and counting have been a key theme for artificial and intelligent biological analysis such as biopsy, drug analysis and decease diagnosis. Along with the rapid development of microfluidics and lab-on-chip technologies, in-vitro live cell analysis has been one of the critical tasks for both research and industry communities. However, it is a great challenge to obtain and then predict the precise information of live cells from numerous microscopic videos and images. In this paper, we investigated in-vitro detection of white blood cells using deep neural networks, and discussed how state-of-the-art machine learning techniques could fulfil the needs of medical diagnosis. The approach we used in this study was based on Faster Region-based Convolutional Neural Networks (Faster RCNNs), and a transfer learning process was applied to apply this technique to the microscopic detection of blood cells. Our experimental results demonstrated that fast and efficient analysis of blood cells via automated microscopic imaging can achieve much better accuracy and faster speed than the conventionally applied methods, implying a promising future of this technology to be applied to the microfluidic point-of-care medical devices.
Tasks	Medical Diagnosis, Transfer Learning
Published	2019-09-11
URL	https://arxiv.org/abs/1909.05393v1
PDF	https://arxiv.org/pdf/1909.05393v1.pdf
PWC	https://paperswithcode.com/paper/automated-blood-cell-detection-and-counting
Repo
Framework

Estimating Local Function Complexity via Mixture of Gaussian Processes


Title	Estimating Local Function Complexity via Mixture of Gaussian Processes
Authors	Danny Panknin, Shinichi Nakajima, Thanh Binh Bui, Klaus-Robert Müller
Abstract	Real world data often exhibit inhomogeneity, e.g., the noise level, the sampling distribution or the complexity of the target function may change over the input space. In this paper, we try to isolate local function complexity in a practical, robust way. This is achieved by first estimating the locally optimal kernel bandwidth as a functional relationship. Specifically, we propose Spatially Adaptive Bandwidth Estimation in Regression (SABER), which employs the mixture of experts consisting of multinomial kernel logistic regression as a gate and Gaussian process regression models as experts. Using the locally optimal kernel bandwidths, we deduce an estimate to the local function complexity by drawing parallels to the theory of locally linear smoothing. We demonstrate the usefulness of local function complexity for model interpretation and active learning in quantum chemistry experiments and fluid dynamics simulations.
Tasks	Active Learning, Gaussian Processes
Published	2019-02-27
URL	https://arxiv.org/abs/1902.10664v3
PDF	https://arxiv.org/pdf/1902.10664v3.pdf
PWC	https://paperswithcode.com/paper/local-bandwidth-estimation-via-mixture-of
Repo
Framework

Identifying DNS-tunneled traffic with predictive models


Title	Identifying DNS-tunneled traffic with predictive models
Authors	Andreas Berg, Daniel Forsberg
Abstract	DNS is a distributed, fault tolerant system that avoids a single point of failure. As such it is an integral part of the internet as we use it today and hence deemed a safe protocol which is let through firewalls and proxies with no or little checks. This can be exploited by malicious agents. Network forensics is effective but struggles due to size of data and manual labour. This paper explores to what extent predictive models can be used to predict network traffic, what protocols are tunneled in the DNS protocol and more specifically whether the predictive performance is enhanced when analyzing DNS-queries and responses together and which feature set that can be used for DNS-tunneled network prediction. The tested protocols are SSH, SFTP and Telnet and the machine learning models used are Multi Layered Perceptron and Random Forests. To train the models we extract the IP Packet length, Name length and Name entropy of both the queries and responses in the DNS traffic. With an experimental research strategy it is empirically shown that the performance of the models increases when training the models on the query and respose pairs rather than using only queries or responses. The accuracy of the models is >83% and reduction in data size when features are extracted is roughly 95%. Our results provides evidence that machine learning is a valuable tool in detecting network protocols in a DNS tunnel and that only an small subset of network traffic is needed to detect this anomaly.
Tasks
Published	2019-06-26
URL	https://arxiv.org/abs/1906.11246v1
PDF	https://arxiv.org/pdf/1906.11246v1.pdf
PWC	https://paperswithcode.com/paper/identifying-dns-tunneled-traffic-with
Repo
Framework

Grassmannian Packings in Neural Networks: Learning with Maximal Subspace Packings for Diversity and Anti-Sparsity


Title	Grassmannian Packings in Neural Networks: Learning with Maximal Subspace Packings for Diversity and Anti-Sparsity
Authors	Dian Ang Yap, Nicholas Roberts, Vinay Uday Prabhu
Abstract	Kernel sparsity (“dying ReLUs”) and lack of diversity are commonly observed in CNN kernels, which decreases model capacity. Drawing inspiration from information theory and wireless communications, we demonstrate the intersection of coding theory and deep learning through the Grassmannian subspace packing problem in CNNs. We propose Grassmannian packings for initial kernel layers to be initialized maximally far apart based on chordal or Fubini-Study distance. Convolutional kernels initialized with Grassmannian packings exhibit diverse features and obtain diverse representations. We show that Grassmannian packings, especially in the initial layers, address kernel sparsity and encourage diversity, while improving classification accuracy across shallow and deep CNNs with better convergence rates.
Tasks
Published	2019-11-18
URL	https://arxiv.org/abs/1911.07418v1
PDF	https://arxiv.org/pdf/1911.07418v1.pdf
PWC	https://paperswithcode.com/paper/grassmannian-packings-in-neural-networks
Repo
Framework

Harnessing the Power of Serverless Runtimes for Large-Scale Optimization


Title	Harnessing the Power of Serverless Runtimes for Large-Scale Optimization
Authors	Arda Aytekin, Mikael Johansson
Abstract	The event-driven and elastic nature of serverless runtimes makes them a very efficient and cost-effective alternative for scaling up computations. So far, they have mostly been used for stateless, data parallel and ephemeral computations. In this work, we propose using serverless runtimes to solve generic, large-scale optimization problems. Specifically, we build a master-worker setup using AWS Lambda as the source of our workers, implement a parallel optimization algorithm to solve a regularized logistic regression problem, and show that relative speedups up to 256 workers and efficiencies above 70% up to 64 workers can be expected. We also identify possible algorithmic and system-level bottlenecks, propose improvements, and discuss the limitations and challenges in realizing these improvements.
Tasks
Published	2019-01-10
URL	http://arxiv.org/abs/1901.03161v1
PDF	http://arxiv.org/pdf/1901.03161v1.pdf
PWC	https://paperswithcode.com/paper/harnessing-the-power-of-serverless-runtimes
Repo
Framework

Assessing the Tolerance of Neural Machine Translation Systems Against Speech Recognition Errors


Title	Assessing the Tolerance of Neural Machine Translation Systems Against Speech Recognition Errors
Authors	Nicholas Ruiz, Mattia Antonino Di Gangi, Nicola Bertoldi, Marcello Federico
Abstract	Machine translation systems are conventionally trained on textual resources that do not model phenomena that occur in spoken language. While the evaluation of neural machine translation systems on textual inputs is actively researched in the literature , little has been discovered about the complexities of translating spoken language data with neural models. We introduce and motivate interesting problems one faces when considering the translation of automatic speech recognition (ASR) outputs on neural machine translation (NMT) systems. We test the robustness of sentence encoding approaches for NMT encoder-decoder modeling, focusing on word-based over byte-pair encoding. We compare the translation of utterances containing ASR errors in state-of-the-art NMT encoder-decoder systems against a strong phrase-based machine translation baseline in order to better understand which phenomena present in ASR outputs are better represented under the NMT framework than approaches that represent translation as a linear model.
Tasks	Machine Translation, Speech Recognition
Published	2019-04-24
URL	http://arxiv.org/abs/1904.10997v1
PDF	http://arxiv.org/pdf/1904.10997v1.pdf
PWC	https://paperswithcode.com/paper/assessing-the-tolerance-of-neural-machine
Repo
Framework