January 29, 2020

3042 words 15 mins read

Paper Group ANR 720

Mortality rate forecasting: can recurrent neural networks beat the Lee-Carter model?. MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation. A Large-scale Varying-view RGB-D Action Dataset for Arbitrary-view Human Action Recognition. English-Czech Systems in WMT19: Document-Level Transformer. A computational linguist …

Mortality rate forecasting: can recurrent neural networks beat the Lee-Carter model?


Title	Mortality rate forecasting: can recurrent neural networks beat the Lee-Carter model?
Authors	Gábor Petneházi, József Gáll
Abstract	This article applies a long short-term memory recurrent neural network to mortality rate forecasting. The model can be trained jointly on the mortality rate history of different countries, ages, and sexes. The RNN-based method seems to outperform the popular Lee-Carter model.
Tasks
Published	2019-09-12
URL	https://arxiv.org/abs/1909.05501v2
PDF	https://arxiv.org/pdf/1909.05501v2.pdf
PWC	https://paperswithcode.com/paper/mortality-rate-forecasting-can-recurrent
Repo
Framework

MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation


Title	MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation
Authors	Tianchen Wang, Jinjun Xiong, Xiaowei Xu, Meng Jiang, Yiyu Shi, Haiyun Yuan, Meiping Huang, Jian Zhuang
Abstract	Cardiac magnetic resonance imaging (MRI) is an essential tool for MRI-guided surgery and real-time intervention. The MRI videos are expected to be segmented on-the-fly in real practice. However, existing segmentation methods would suffer from drastic accuracy loss when modified for speedup. In this work, we propose Multiscale Statistical U-Net (MSU-Net) for real-time 3D MRI video segmentation in cardiac surgical guidance. Our idea is to model the input samples as multiscale canonical form distributions for speedup, while the spatio-temporal correlation is still fully utilized. A parallel statistical U-Net is then designed to efficiently process these distributions. The fast data sampling and efficient parallel structure of MSU-Net endorse the fast and accurate inference. Compared with vanilla U-Net and a modified state-of-the-art method GridNet, our method achieves up to 268% and 237% speedup with 1.6% and 3.6% increased Dice scores.
Tasks	Video Semantic Segmentation
Published	2019-09-15
URL	https://arxiv.org/abs/1909.06726v1
PDF	https://arxiv.org/pdf/1909.06726v1.pdf
PWC	https://paperswithcode.com/paper/msu-net-multiscale-statistical-u-net-for-real
Repo
Framework

A Large-scale Varying-view RGB-D Action Dataset for Arbitrary-view Human Action Recognition


Title	A Large-scale Varying-view RGB-D Action Dataset for Arbitrary-view Human Action Recognition
Authors	Yanli Ji, Feixiang Xu, Yang Yang, Fumin Shen, Heng Tao Shen, Wei-Shi Zheng
Abstract	Current researches of action recognition mainly focus on single-view and multi-view recognition, which can hardly satisfies the requirements of human-robot interaction (HRI) applications to recognize actions from arbitrary views. The lack of datasets also sets up barriers. To provide data for arbitrary-view action recognition, we newly collect a large-scale RGB-D action dataset for arbitrary-view action analysis, including RGB videos, depth and skeleton sequences. The dataset includes action samples captured in 8 fixed viewpoints and varying-view sequences which covers the entire 360 degree view angles. In total, 118 persons are invited to act 40 action categories, and 25,600 video samples are collected. Our dataset involves more participants, more viewpoints and a large number of samples. More importantly, it is the first dataset containing the entire 360 degree varying-view sequences. The dataset provides sufficient data for multi-view, cross-view and arbitrary-view action analysis. Besides, we propose a View-guided Skeleton CNN (VS-CNN) to tackle the problem of arbitrary-view action recognition. Experiment results show that the VS-CNN achieves superior performance.
Tasks	Skeleton Based Action Recognition, Temporal Action Localization
Published	2019-04-24
URL	http://arxiv.org/abs/1904.10681v1
PDF	http://arxiv.org/pdf/1904.10681v1.pdf
PWC	https://paperswithcode.com/paper/a-large-scale-varying-view-rgb-d-action
Repo
Framework

English-Czech Systems in WMT19: Document-Level Transformer


Title	English-Czech Systems in WMT19: Document-Level Transformer
Authors	Martin Popel, Dominik Macháček, Michal Auersperger, Ondřej Bojar, Pavel Pecina
Abstract	We describe our NMT systems submitted to the WMT19 shared task in English-Czech news translation. Our systems are based on the Transformer model implemented in either Tensor2Tensor (T2T) or Marian framework. We aimed at improving the adequacy and coherence of translated documents by enlarging the context of the source and target. Instead of translating each sentence independently, we split the document into possibly overlapping multi-sentence segments. In case of the T2T implementation, this “document-level”-trained system achieves a $+0.6$ BLEU improvement ($p<0.05$) relative to the same system applied on isolated sentences. To assess the potential effect document-level models might have on lexical coherence, we performed a semi-automatic analysis, which revealed only a few sentences improved in this aspect. Thus, we cannot draw any conclusions from this weak evidence.
Tasks
Published	2019-07-30
URL	https://arxiv.org/abs/1907.12750v1
PDF	https://arxiv.org/pdf/1907.12750v1.pdf
PWC	https://paperswithcode.com/paper/english-czech-systems-in-wmt19-document-level
Repo
Framework

A computational linguistic study of personal recovery in bipolar disorder


Title	A computational linguistic study of personal recovery in bipolar disorder
Authors	Glorianna Jagfeld
Abstract	Mental health research can benefit increasingly fruitfully from computational linguistics methods, given the abundant availability of language data in the internet and advances of computational tools. This interdisciplinary project will collect and analyse social media data of individuals diagnosed with bipolar disorder with regard to their recovery experiences. Personal recovery - living a satisfying and contributing life along symptoms of severe mental health issues - so far has only been investigated qualitatively with structured interviews and quantitatively with standardised questionnaires with mainly English-speaking participants in Western countries. Complementary to this evidence, computational linguistic methods allow us to analyse first-person accounts shared online in large quantities, representing unstructured settings and a more heterogeneous, multilingual population, to draw a more complete picture of the aspects and mechanisms of personal recovery in bipolar disorder.
Tasks
Published	2019-06-03
URL	https://arxiv.org/abs/1906.01010v1
PDF	https://arxiv.org/pdf/1906.01010v1.pdf
PWC	https://paperswithcode.com/paper/a-computational-linguistic-study-of-personal
Repo
Framework

Adversarial Domain Adaptation Being Aware of Class Relationships


Title	Adversarial Domain Adaptation Being Aware of Class Relationships
Authors	Zeya Wang, Baoyu Jing, Yang Ni, Nanqing Dong, Pengtao Xie, Eric P. Xing
Abstract	Adversarial training is a useful approach to promote the learning of transferable representations across the source and target domains, which has been widely applied for domain adaptation (DA) tasks based on deep neural networks. Until very recently, existing adversarial domain adaptation (ADA) methods ignore the useful information from the label space, which is an important factor accountable for the complicated data distributions associated with different semantic classes. Especially, the inter-class semantic relationships have been rarely considered and discussed in the current work of transfer learning. In this paper, we propose a novel relationship-aware adversarial domain adaptation (RADA) algorithm, which first utilizes a single multi-class domain discriminator to enforce the learning of inter-class dependency structure during domain-adversarial training and then aligns this structure with the inter-class dependencies that are characterized from training the label predictor on source domain. Specifically, we impose a regularization term to penalize the structure discrepancy between the inter-class dependencies respectively estimated from domain discriminator and label predictor. Through this alignment, our proposed method makes the adversarial domain adaptation aware of the class relationships. Empirical studies show that the incorporation of class relationships significantly improves the performance on benchmark datasets.
Tasks	Domain Adaptation, Transfer Learning
Published	2019-05-28
URL	https://arxiv.org/abs/1905.11931v2
PDF	https://arxiv.org/pdf/1905.11931v2.pdf
PWC	https://paperswithcode.com/paper/adversarial-domain-adaptation-being-aware-of
Repo
Framework

Style Conditioned Recommendations


Title	Style Conditioned Recommendations
Authors	Murium Iqbal, Kamelia Aryafar, Timothy Anderton
Abstract	We propose Style Conditioned Recommendations (SCR) and introduce style injection as a method to diversify recommendations. We use Conditional Variational Autoencoder (CVAE) architecture, where both the encoder and decoder are conditioned on a user profile learned from item content data. This allows us to apply style transfer methodologies to the task of recommendations, which we refer to as injection. To enable style injection, user profiles are learned to be interpretable such that they express users’ propensities for specific predefined styles. These are learned via label-propagation from a dataset of item content, with limited labeled points. To perform injection, the condition on the encoder is learned while the condition on the decoder is selected per explicit feedback. Explicit feedback can be taken either from a user’s response to a style or interest quiz, or from item ratings. In the absence of explicit feedback, the condition at the encoder is applied to the decoder. We show a 12% improvement on NDCG@20 over the traditional VAE based approach and an average 22% improvement on AUC across all classes for predicting user style profiles against our best performing baseline. After injecting styles we compare the user style profile to the style of the recommendations and show that injected styles have an average +133% increase in presence. Our results show that style injection is a powerful method to diversify recommendations while maintaining personal relevance. Our main contribution is an application of a semi-supervised approach that extends item labels to interpretable user profiles.
Tasks	Style Transfer
Published	2019-07-25
URL	https://arxiv.org/abs/1907.12388v2
PDF	https://arxiv.org/pdf/1907.12388v2.pdf
PWC	https://paperswithcode.com/paper/style-conditioned-recommendations
Repo
Framework

Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentation


Title	Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentation
Authors	Yang He, Shadi Rahimian, Bernt Schiele, Mario Fritz
Abstract	Today’s success of state of the art methods for semantic segmentation is driven by large datasets. Data is considered an important asset that needs to be protected, as the collection and annotation of such datasets comes at significant efforts and associated costs. In addition, visual data might contain private or sensitive information, that makes it equally unsuited for public release. Unfortunately, recent work on membership inference in the broader area of adversarial machine learning and inference attacks on machine learning models has shown that even black box classifiers leak information on the dataset that they were trained on. We present the first attacks and defenses for complex, state of the art models for semantic segmentation. In order to mitigate the associated risks, we also study a series of defenses against such membership inference attacks and find effective counter measures against the existing risks. Finally, we extensively evaluate our attacks and defenses on a range of relevant real-world datasets: Cityscapes, BDD100K, and Mapillary Vistas.
Tasks	Semantic Segmentation
Published	2019-12-20
URL	https://arxiv.org/abs/1912.09685v1
PDF	https://arxiv.org/pdf/1912.09685v1.pdf
PWC	https://paperswithcode.com/paper/segmentations-leak-membership-inference
Repo
Framework

Quantitative Overfitting Management for Human-in-the-loop ML Application Development with ease.ml/meter


Title	Quantitative Overfitting Management for Human-in-the-loop ML Application Development with ease.ml/meter
Authors	Frances Ann Hubis, Wentao Wu, Ce Zhang
Abstract	Simplifying machine learning (ML) application development, including distributed computation, programming interface, resource management, model selection, etc, has attracted intensive interests recently. These research efforts have significantly improved the efficiency and the degree of automation of developing ML models. In this paper, we take a first step in an orthogonal direction towards automated quality management for human-in-the-loop ML application development. We build ease. ml/meter, a system that can automatically detect and measure the degree of overfitting during the whole lifecycle of ML application development. ease. ml/meter returns overfitting signals with strong probabilistic guarantees, based on which developers can take appropriate actions. In particular, ease. ml/meter provides principled guidelines to simple yet nontrivial questions regarding desired validation and test data sizes, which are among commonest questions raised by developers. The fact that ML application development is typically a continuous procedure further worsens the situation: The validation and test data sets can lose their statistical power quickly due to multiple accesses, especially in the presence of adaptive analysis. ease. ml/meter addresses these challenges by leveraging a collection of novel techniques and optimizations, resulting in practically tractable data sizes without compromising the probabilistic guarantees. We present the design and implementation details of ease. ml/meter, as well as detailed theoretical analysis and empirical evaluation of its effectiveness.
Tasks	Model Selection
Published	2019-06-01
URL	https://arxiv.org/abs/1906.00299v3
PDF	https://arxiv.org/pdf/1906.00299v3.pdf
PWC	https://paperswithcode.com/paper/190600299
Repo
Framework

Multi-Label Network Classification via Weighted Personalized Factorizations


Title	Multi-Label Network Classification via Weighted Personalized Factorizations
Authors	Ahmed Rashed, Josif Grabocka, Lars Schmidt-Thieme
Abstract	Multi-label network classification is a well-known task that is being used in a wide variety of web-based and non-web-based domains. It can be formalized as a multi-relational learning task for predicting nodes labels based on their relations within the network. In sparse networks, this prediction task can be very challenging when only implicit feedback information is available such as in predicting user interests in social networks. Current approaches rely on learning per-node latent representations by utilizing the network structure, however, implicit feedback relations are naturally sparse and contain only positive observed feedbacks which mean that these approaches will treat all observed relations as equally important. This is not necessarily the case in real-world scenarios as implicit relations might have semantic weights which reflect the strength of those relations. If those weights can be approximated, the models can be trained to differentiate between strong and weak relations. In this paper, we propose a weighted personalized two-stage multi-relational matrix factorization model with Bayesian personalized ranking loss for network classification that utilizes basic transitive node similarity function for weighting implicit feedback relations. Experiments show that the proposed model significantly outperforms the state-of-art models on three different real-world web-based datasets and a biology-based dataset.
Tasks	Relational Reasoning
Published	2019-02-25
URL	http://arxiv.org/abs/1902.09294v1
PDF	http://arxiv.org/pdf/1902.09294v1.pdf
PWC	https://paperswithcode.com/paper/multi-label-network-classification-via
Repo
Framework

Asymptotically Unambitious Artificial General Intelligence


Title	Asymptotically Unambitious Artificial General Intelligence
Authors	Michael K Cohen, Badri Vellambi, Marcus Hutter
Abstract	General intelligence, the ability to solve arbitrary solvable problems, is supposed by many to be artificially constructible. Narrow intelligence, the ability to solve a given particularly difficult problem, has seen impressive recent development. Notable examples include self-driving cars, Go engines, image classifiers, and translators. Artificial General Intelligence (AGI) presents dangers that narrow intelligence does not: if something smarter than us across every domain were indifferent to our concerns, it would be an existential threat to humanity, just as we threaten many species despite no ill will. Even the theory of how to maintain the alignment of an AGI’s goals with our own has proven highly elusive. We present the first algorithm we are aware of for asymptotically unambitious AGI, where “unambitiousness” includes not seeking arbitrary power. Thus, we identify an exception to the Instrumental Convergence Thesis, which is roughly that by default, an AGI would seek power, including over us.
Tasks	Self-Driving Cars
Published	2019-05-29
URL	https://arxiv.org/abs/1905.12186v3
PDF	https://arxiv.org/pdf/1905.12186v3.pdf
PWC	https://paperswithcode.com/paper/asymptotically-unambitious-artificial-general
Repo
Framework

Understanding Pedestrian-Vehicle Interactions with Vehicle Mounted Vision: An LSTM Model and Empirical Analysis


Title	Understanding Pedestrian-Vehicle Interactions with Vehicle Mounted Vision: An LSTM Model and Empirical Analysis
Authors	Daniela A. Ridel, Nachiket Deo, Denis Wolf, Mohan M. Trivedi
Abstract	Pedestrians and vehicles often share the road in complex inner city traffic. This leads to interactions between the vehicle and pedestrians, with each affecting the other’s motion. In order to create robust methods to reason about pedestrian behavior and to design interfaces of communication between self-driving cars and pedestrians we need to better understand such interactions. In this paper, we present a data-driven approach to implicitly model pedestrians’ interactions with vehicles, to better predict pedestrian behavior. We propose a LSTM model that takes as input the past trajectories of the pedestrian and ego-vehicle, and pedestrian head orientation, and predicts the future positions of the pedestrian. Our experiments based on a real-world, inner city dataset captured with vehicle mounted cameras, show that the usage of such cues improve pedestrian prediction when compared to a baseline that purely uses the past trajectory of the pedestrian.
Tasks	Self-Driving Cars
Published	2019-05-14
URL	https://arxiv.org/abs/1905.05350v1
PDF	https://arxiv.org/pdf/1905.05350v1.pdf
PWC	https://paperswithcode.com/paper/understanding-pedestrian-vehicle-interactions
Repo
Framework

Machinic Surrogates: Human-Machine Relationships in Computational Creativity


Title	Machinic Surrogates: Human-Machine Relationships in Computational Creativity
Authors	Ardavan Bidgoli, Eunsu Kang, Daniel Cardoso Llach
Abstract	Recent advancements in artificial intelligence (AI) and its sub-branch machine learning (ML) promise machines that go beyond the boundaries of automation and behave autonomously. Applications of these machines in creative practices such as art and design entail relationships between users and machines that have been described as a form of collaboration or co-creation between computational and human agents. This paper uses examples from art and design to argue that this frame is incomplete as it fails to acknowledge the socio-technical nature of AI systems, and the different human agencies involved in their design, implementation, and operation. Situating applications of AI-enabled tools in creative practices in a spectrum between automation and autonomy, this paper distinguishes different kinds of human engagement elicited by systems deemed automated or autonomous. Reviewing models of artistic collaboration during the late 20th century, it suggests that collaboration is at the core of these artistic practices. We build upon the growing literature of machine learning and art to look for the human agencies inscribed in works of computational creativity, and expand the co-creation frame to incorporate emerging forms of human-human collaboration mediated through technical artifacts such as algorithms and data.
Tasks
Published	2019-08-03
URL	https://arxiv.org/abs/1908.01133v1
PDF	https://arxiv.org/pdf/1908.01133v1.pdf
PWC	https://paperswithcode.com/paper/machinic-surrogates-human-machine
Repo
Framework


Title	Uncovering Dominant Social Class in Neighborhoods through Building Footprints: A Case Study of Residential Zones in Massachusetts using Computer Vision
Authors	Qianhui Liang, Zhoutong Wang
Abstract	In urban theory, urban form is related to social and economic status. This paper explores to uncover zip-code level income through urban form by analyzing figure-ground map, a simple, prevailing and precise representation of urban form in the field of urban study. Deep learning in computer vision enables such representation maps to be studied at a large scale. We propose to train a DCNN model to identify and uncover the internal bridge between social class and urban form. Further, using hand-crafted informative visual features related with urban form properties (building size, building density, etc.), we apply a random forest classifier to interpret how morphological properties are related with social class.
Tasks
Published	2019-06-12
URL	https://arxiv.org/abs/1906.05352v1
PDF	https://arxiv.org/pdf/1906.05352v1.pdf
PWC	https://paperswithcode.com/paper/uncovering-dominant-social-class-in
Repo
Framework

Lexicase selection in Learning Classifier Systems


Title	Lexicase selection in Learning Classifier Systems
Authors	Sneha Aenugu, Lee Spector
Abstract	The lexicase parent selection method selects parents by considering performance on individual data points in random order instead of using a fitness function based on an aggregated data accuracy. While the method has demonstrated promise in genetic programming and more recently in genetic algorithms, its applications in other forms of evolutionary machine learning have not been explored. In this paper, we investigate the use of lexicase parent selection in Learning Classifier Systems (LCS) and study its effect on classification problems in a supervised setting. We further introduce a new variant of lexicase selection, called batch-lexicase selection, which allows for the tuning of selection pressure. We compare the two lexicase selection methods with tournament and fitness proportionate selection methods on binary classification problems. We show that batch-lexicase selection results in the creation of more generic rules which is favorable for generalization on future data. We further show that batch-lexicase selection results in better generalization in situations of partial or missing data.
Tasks
Published	2019-07-10
URL	https://arxiv.org/abs/1907.04736v1
PDF	https://arxiv.org/pdf/1907.04736v1.pdf
PWC	https://paperswithcode.com/paper/lexicase-selection-in-learning-classifier
Repo
Framework