February 1, 2020

3105 words 15 mins read

Paper Group AWR 277

Avoiding Resentment Via Monotonic Fairness. Attentive Mimicking: Better Word Embeddings by Attending to Informative Contexts. How Large a Vocabulary Does Text Classification Need? A Variational Approach to Vocabulary Selection. 12-in-1: Multi-Task Vision and Language Representation Learning. Reducing Domain Gap via Style-Agnostic Networks. Semantic …

Avoiding Resentment Via Monotonic Fairness


Title	Avoiding Resentment Via Monotonic Fairness
Authors	Guy W. Cole, Sinead A. Williamson
Abstract	Classifiers that achieve demographic balance by explicitly using protected attributes such as race or gender are often politically or culturally controversial due to their lack of individual fairness, i.e. individuals with similar qualifications will receive different outcomes. Individually and group fair decision criteria can produce counter-intuitive results, e.g. that the optimal constrained boundary may reject intuitively better candidates due to demographic imbalance in similar candidates. Both approaches can be seen as introducing individual resentment, where some individuals would have received a better outcome if they either belonged to a different demographic class and had the same qualifications, or if they remained in the same class but had objectively worse qualifications (e.g. lower test scores). We show that both forms of resentment can be avoided by using monotonically constrained machine learning models to create individually fair, demographically balanced classifiers.
Tasks
Published	2019-09-03
URL	https://arxiv.org/abs/1909.01251v1
PDF	https://arxiv.org/pdf/1909.01251v1.pdf
PWC	https://paperswithcode.com/paper/avoiding-resentment-via-monotonic-fairness
Repo	https://github.com/throwaway20190523/MonotonicFairness
Framework	tf

Attentive Mimicking: Better Word Embeddings by Attending to Informative Contexts


Title	Attentive Mimicking: Better Word Embeddings by Attending to Informative Contexts
Authors	Timo Schick, Hinrich Schütze
Abstract	Learning high-quality embeddings for rare words is a hard problem because of sparse context information. Mimicking (Pinter et al., 2017) has been proposed as a solution: given embeddings learned by a standard algorithm, a model is first trained to reproduce embeddings of frequent words from their surface form and then used to compute embeddings for rare words. In this paper, we introduce attentive mimicking: the mimicking model is given access not only to a word’s surface form, but also to all available contexts and learns to attend to the most informative and reliable contexts for computing an embedding. In an evaluation on four tasks, we show that attentive mimicking outperforms previous work for both rare and medium-frequency words. Thus, compared to previous work, attentive mimicking improves embeddings for a much larger part of the vocabulary, including the medium-frequency range.
Tasks	Word Embeddings
Published	2019-04-02
URL	http://arxiv.org/abs/1904.01617v2
PDF	http://arxiv.org/pdf/1904.01617v2.pdf
PWC	https://paperswithcode.com/paper/attentive-mimicking-better-word-embeddings-by
Repo	https://github.com/timoschick/form-context-model
Framework	none

How Large a Vocabulary Does Text Classification Need? A Variational Approach to Vocabulary Selection


Title	How Large a Vocabulary Does Text Classification Need? A Variational Approach to Vocabulary Selection
Authors	Wenhu Chen, Yu Su, Yilin Shen, Zhiyu Chen, Xifeng Yan, William Wang
Abstract	With the rapid development in deep learning, deep neural networks have been widely adopted in many real-life natural language applications. Under deep neural networks, a pre-defined vocabulary is required to vectorize text inputs. The canonical approach to select pre-defined vocabulary is based on the word frequency, where a threshold is selected to cut off the long tail distribution. However, we observed that such simple approach could easily lead to under-sized vocabulary or over-sized vocabulary issues. Therefore, we are interested in understanding how the end-task classification accuracy is related to the vocabulary size and what is the minimum required vocabulary size to achieve a specific performance. In this paper, we provide a more sophisticated variational vocabulary dropout (VVD) based on variational dropout to perform vocabulary selection, which can intelligently select the subset of the vocabulary to achieve the required performance. To evaluate different algorithms on the newly proposed vocabulary selection problem, we propose two new metrics: Area Under Accuracy-Vocab Curve and Vocab Size under X% Accuracy Drop. Through extensive experiments on various NLP classification tasks, our variational framework is shown to significantly outperform the frequency-based and other selection baselines on these metrics.
Tasks	Text Classification
Published	2019-02-27
URL	http://arxiv.org/abs/1902.10339v4
PDF	http://arxiv.org/pdf/1902.10339v4.pdf
PWC	https://paperswithcode.com/paper/how-large-a-vocabulary-does-text
Repo	https://github.com/wenhuchen/Variational-Vocabulary-Selection
Framework	tf

12-in-1: Multi-Task Vision and Language Representation Learning


Title	12-in-1: Multi-Task Vision and Language Representation Learning
Authors	Jiasen Lu, Vedanuj Goswami, Marcus Rohrbach, Devi Parikh, Stefan Lee
Abstract	Much of vision-and-language research focuses on a small but diverse set of independent tasks and supporting datasets often studied in isolation; however, the visually-grounded language understanding skills required for success at these tasks overlap significantly. In this work, we investigate these relationships between vision-and-language tasks by developing a large-scale, multi-task training regime. Our approach culminates in a single model on 12 datasets from four broad categories of task including visual question answering, caption-based image retrieval, grounding referring expressions, and multi-modal verification. Compared to independently trained single-task models, this represents a reduction from approximately 3 billion parameters to 270 million while simultaneously improving performance by 2.05 points on average across tasks. We use our multi-task framework to perform in-depth analysis of the effect of joint training diverse tasks. Further, we show that finetuning task-specific models from our single multi-task model can lead to further improvements, achieving performance at or above the state-of-the-art.
Tasks	Image Retrieval, Question Answering, Representation Learning, Visual Question Answering
Published	2019-12-05
URL	https://arxiv.org/abs/1912.02315v1
PDF	https://arxiv.org/pdf/1912.02315v1.pdf
PWC	https://paperswithcode.com/paper/12-in-1-multi-task-vision-and-language
Repo	https://github.com/jiasenlu/vilbert_beta
Framework	pytorch

Reducing Domain Gap via Style-Agnostic Networks


Title	Reducing Domain Gap via Style-Agnostic Networks
Authors	Hyeonseob Nam, HyunJae Lee, Jongchan Park, Wonjun Yoon, Donggeun Yoo
Abstract	Convolutional Neural Networks (CNNs) often fail to maintain their performance when they confront new test domains. This has posed substantial obstacles to real-world applications of deep learning. Recent studies suggest that one of the main causes of this problem is CNNs’ inductive bias towards image styles (i.e. textures), which are highly dependent on domains, rather than contents (i.e. shapes). Motivated by this, we propose Style-Agnostic Networks (SagNets) which mitigate the style bias to generalize better under domain shift. Our experiments demonstrate that SagNets successfully reduce the style bias as well as domain discrepancy, and clarify a strong correlation between the bias and domain gap. Finally, SagNets achieve remarkable performance improvements in a wide range of cross-domain tasks, including domain generalization, unsupervised domain adaptation, and semi-supervised domain adaptation.
Tasks	Domain Adaptation, Domain Generalization, Unsupervised Domain Adaptation
Published	2019-10-25
URL	https://arxiv.org/abs/1910.11645v2
PDF	https://arxiv.org/pdf/1910.11645v2.pdf
PWC	https://paperswithcode.com/paper/reducing-domain-gap-via-style-agnostic
Repo	https://github.com/hyeonseobnam/style-agnostic-networks
Framework	pytorch

Semantic Guided Single Image Reflection Removal


Title	Semantic Guided Single Image Reflection Removal
Authors	Yunfei Liu, Yu Li, Shaodi You, Feng Lu
Abstract	Reflection is common in images capturing scenes behind a glass window, which is not only a disturbance visually but also influence the performance of other computer vision algorithms. Single image reflection removal is an ill-posed problem because the color at each pixel needs to be separated into two values, i.e., the desired clear background and the reflection. To solve it, existing methods propose priors such as smoothness, color consistency. However, the low-level priors are not reliable in complex scenes, for instance, when capturing a real outdoor scene through a window, both the foreground and background contain both smooth and sharp area and a variety of color. In this paper, inspired by the fact that human can separate the two layers easily by recognizing the objects, we use the object semantic as guidance to force the same semantic object belong to the same layer. Extensive experiments on different datasets show that adding the semantic information offers a significant improvement to reflection separation. We also demonstrate the applications of the proposed method to other computer vision tasks.
Tasks
Published	2019-07-27
URL	https://arxiv.org/abs/1907.11912v2
PDF	https://arxiv.org/pdf/1907.11912v2.pdf
PWC	https://paperswithcode.com/paper/semantic-guided-single-image-reflection
Repo	https://github.com/DreamtaleCore/SGRRN
Framework	pytorch

DeepSphere: towards an equivariant graph-based spherical CNN


Title	DeepSphere: towards an equivariant graph-based spherical CNN
Authors	Michaël Defferrard, Nathanaël Perraudin, Tomasz Kacprzak, Raphael Sgier
Abstract	Spherical data is found in many applications. By modeling the discretized sphere as a graph, we can accommodate non-uniformly distributed, partial, and changing samplings. Moreover, graph convolutions are computationally more efficient than spherical convolutions. As equivariance is desired to exploit rotational symmetries, we discuss how to approach rotation equivariance using the graph neural network introduced in Defferrard et al. (2016). Experiments show good performance on rotation-invariant learning problems. Code and examples are available at https://github.com/SwissDataScienceCenter/DeepSphere
Tasks
Published	2019-04-08
URL	http://arxiv.org/abs/1904.05146v1
PDF	http://arxiv.org/pdf/1904.05146v1.pdf
PWC	https://paperswithcode.com/paper/deepsphere-towards-an-equivariant-graph-based
Repo	https://github.com/SwissDataScienceCenter/DeepSphere
Framework	tf

Domain-Specific Batch Normalization for Unsupervised Domain Adaptation


Title	Domain-Specific Batch Normalization for Unsupervised Domain Adaptation
Authors	Woong-Gi Chang, Tackgeun You, Seonguk Seo, Suha Kwak, Bohyung Han
Abstract	We propose a novel unsupervised domain adaptation framework based on domain-specific batch normalization in deep neural networks. We aim to adapt to both domains by specializing batch normalization layers in convolutional neural networks while allowing them to share all other model parameters, which is realized by a two-stage algorithm. In the first stage, we estimate pseudo-labels for the examples in the target domain using an external unsupervised domain adaptation algorithm—for example, MSTN or CPUA—integrating the proposed domain-specific batch normalization. The second stage learns the final models using a multi-task classification loss for the source and target domains. Note that the two domains have separate batch normalization layers in both stages. Our framework can be easily incorporated into the domain adaptation techniques based on deep neural networks with batch normalization layers. We also present that our approach can be extended to the problem with multiple source domains. The proposed algorithm is evaluated on multiple benchmark datasets and achieves the state-of-the-art accuracy in the standard setting and the multi-source domain adaption scenario.
Tasks	Domain Adaptation, Unsupervised Domain Adaptation
Published	2019-05-27
URL	https://arxiv.org/abs/1906.03950v1
PDF	https://arxiv.org/pdf/1906.03950v1.pdf
PWC	https://paperswithcode.com/paper/domain-specific-batch-normalization-for-1
Repo	https://github.com/wgchang/DSBN
Framework	pytorch

Physics-Informed Neural Networks for Multiphysics Data Assimilation with Application to Subsurface Transport


Title	Physics-Informed Neural Networks for Multiphysics Data Assimilation with Application to Subsurface Transport
Authors	QiZhi He, David Brajas-Solano, Guzel Tartakovsky, Alexandre M. Tartakovsky
Abstract	Data assimilation for parameter and state estimation in subsurface transport problems remains a significant challenge due to the sparsity of measurements, the heterogeneity of porous media, and the high computational cost of forward numerical models. We present a physics-informed deep neural networks (DNNs) machine learning method for estimating space-dependent hydraulic conductivity, hydraulic head, and concentration fields from sparse measurements. In this approach, we employ individual DNNs to approximate the unknown parameters (e.g., hydraulic conductivity) and states (e.g., hydraulic head and concentration) of a physical system, and jointly train these DNNs by minimizing the loss function that consists of the governing equations residuals in addition to the error with respect to measurement data. We apply this approach to assimilate conductivity, hydraulic head, and concentration measurements for joint inversion of the conductivity, hydraulic head, and concentration fields in a steady-state advection–dispersion problem. We study the accuracy of the physics-informed DNN approach with respect to data size, number of variables (conductivity and head versus conductivity, head, and concentration), DNNs size, and DNN initialization during training. We demonstrate that the physics-informed DNNs are significantly more accurate than standard data-driven DNNs when the training set consists of sparse data. We also show that the accuracy of parameter estimation increases as additional variables are inverted jointly.
Tasks
Published	2019-12-06
URL	https://arxiv.org/abs/1912.02968v1
PDF	https://arxiv.org/pdf/1912.02968v1.pdf
PWC	https://paperswithcode.com/paper/physics-informed-neural-networks-for
Repo	https://github.com/qzhe-mechanics/DataAssi-transport
Framework	none

Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning


Title	Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning
Authors	Cameron Voloshin, Hoang M. Le, Nan Jiang, Yisong Yue
Abstract	The disparate experimental conditions in recent off-policy policy evaluation (OPE) literature make it difficult both for practitioners to choose a reliable estimator for their application domain, as well as for researchers to identify fruitful research directions. In this work, we present the first detailed empirical study of a broad suite of OPE methods. Based on thousands of experiments and empirical analysis, we offer a summarized set of guidelines to advance the understanding of OPE performance in practice, and suggest directions for future research. Along the way, our empirical findings challenge several commonly held beliefs about which class of approaches tends to perform well. Our accompanying software implementation serves as a first comprehensive benchmark for OPE.
Tasks
Published	2019-11-15
URL	https://arxiv.org/abs/1911.06854v2
PDF	https://arxiv.org/pdf/1911.06854v2.pdf
PWC	https://paperswithcode.com/paper/empirical-study-of-off-policy-policy
Repo	https://github.com/clvoloshin/OPE-tools
Framework	none

I-Keyboard: Fully Imaginary Keyboard on Touch Devices Empowered by Deep Neural Decoder


Title	I-Keyboard: Fully Imaginary Keyboard on Touch Devices Empowered by Deep Neural Decoder
Authors	Ue-Hwan Kim, Sahng-Min Yoo, Jong-Hwan Kim
Abstract	Text-entry aims to provide an effective and efficient pathway for humans to deliver their messages to computers. With the advent of mobile computing, the recent focus of text-entry research has moved from physical keyboards to soft keyboards. Current soft keyboards, however, increase the typo rate due to lack of tactile feedback and degrade the usability of mobile devices due to their large portion on screens. To tackle these limitations, we propose a fully imaginary keyboard (I-Keyboard) with a deep neural decoder (DND). The invisibility of I-Keyboard maximizes the usability of mobile devices and DND empowered by a deep neural architecture allows users to start typing from any position on the touch screens at any angle. To the best of our knowledge, the eyes-free ten-finger typing scenario of I-Keyboard which does not necessitate both a calibration step and a predefined region for typing is first explored in this work. For the purpose of training DND, we collected the largest user data in the process of developing I-Keyboard. We verified the performance of the proposed I-Keyboard and DND by conducting a series of comprehensive simulations and experiments under various conditions. I-Keyboard showed 18.95% and 4.06% increases in typing speed (45.57 WPM) and accuracy (95.84%), respectively over the baseline.
Tasks	Calibration
Published	2019-07-31
URL	https://arxiv.org/abs/1907.13285v1
PDF	https://arxiv.org/pdf/1907.13285v1.pdf
PWC	https://paperswithcode.com/paper/i-keyboard-fully-imaginary-keyboard-on-touch
Repo	https://github.com/Uehwan/I-Keyboard
Framework	tf

CLEAR: A Consistent Lifting, Embedding, and Alignment Rectification Algorithm for Multi-View Data Association


Title	CLEAR: A Consistent Lifting, Embedding, and Alignment Rectification Algorithm for Multi-View Data Association
Authors	Kaveh Fathian, Kasra Khosoussi, Yulun Tian, Parker Lusk, Jonathan P. How
Abstract	Many robotics applications require alignment and fusion of observations obtained at multiple views to form a global model of the environment. Multi-way data association methods provide a mechanism to improve alignment accuracy of pairwise associations and ensure their consistency. However, existing methods that solve this computationally challenging problem are often too slow for real-time applications. Furthermore, some of the existing techniques can violate the cycle consistency principle, thus drastically reducing the fusion accuracy. This work presents the CLEAR (Consistent Lifting, Embedding, and Alignment Rectification) algorithm to address these issues. By leveraging insights from the multi-way matching and spectral graph clustering literature, CLEAR provides cycle consistent and accurate solutions in a computationally efficient manner. Numerical experiments on both synthetic and real datasets are carried out to demonstrate the scalability and superior performance of our algorithm in real-world problems. This algorithmic framework can provide significant improvement in the accuracy and efficiency of existing discrete assignment problems, which traditionally use pairwise (but potentially inconsistent) correspondences. An implementation of CLEAR is made publicly available online.
Tasks	Graph Clustering, Spectral Graph Clustering
Published	2019-02-06
URL	https://arxiv.org/abs/1902.02256v3
PDF	https://arxiv.org/pdf/1902.02256v3.pdf
PWC	https://paperswithcode.com/paper/clear-a-consistent-lifting-embedding-and
Repo	https://github.com/mit-acl/clear
Framework	none

Calibration of Asynchronous Camera Networks: CALICO


Title	Calibration of Asynchronous Camera Networks: CALICO
Authors	Amy Tabb, Henry Medeiros, Mitchell J. Feldmann, Thiago T. Santos
Abstract	Camera network and multi-camera calibration for external parameters is a necessary step for a variety of contexts in computer vision and robotics, ranging from three-dimensional reconstruction to human activity tracking. This paper describes CALICO, a method for camera network and/or multi-camera calibration suitable for challenging contexts: the cameras may not share a common field of view and the network may be asynchronous. The calibration object required is one or more rigidly attached planar calibration patterns, which are distinguishable from one another, such as aruco or charuco patterns. We formulate the camera network and/or multi-camera calibration problem using rigidity constraints, represented as a system of equations, and an approximate solution is found through a two-step process. Simulated and real experiments, including an asynchronous camera network, multicamera system, and rotating imaging system, demonstrate the method in a variety of settings. Median reconstruction accuracy error was less than $0.41$ mm$^2$ for all datasets. This method is suitable for novice users to calibrate a camera network, and the modularity of the calibration object also allows for disassembly, shipping, and the use of this method in a variety of large and small spaces.
Tasks	Calibration, Object Reconstruction
Published	2019-03-15
URL	https://arxiv.org/abs/1903.06811v2
PDF	https://arxiv.org/pdf/1903.06811v2.pdf
PWC	https://paperswithcode.com/paper/calibration-of-asynchronous-camera-networks
Repo	https://github.com/amy-tabb/calico
Framework	none


Title	Graph Neural Networks for Social Recommendation
Authors	Wenqi Fan, Yao Ma, Qing Li, Yuan He, Eric Zhao, Jiliang Tang, Dawei Yin
Abstract	In recent years, Graph Neural Networks (GNNs), which can naturally integrate node information and topological structure, have been demonstrated to be powerful in learning on graph data. These advantages of GNNs provide great potential to advance social recommendation since data in social recommender systems can be represented as user-user social graph and user-item graph; and learning latent factors of users and items is the key. However, building social recommender systems based on GNNs faces challenges. For example, the user-item graph encodes both interactions and their associated opinions; social relations have heterogeneous strengths; users involve in two graphs (e.g., the user-user social graph and the user-item graph). To address the three aforementioned challenges simultaneously, in this paper, we present a novel graph neural network framework (GraphRec) for social recommendations. In particular, we provide a principled approach to jointly capture interactions and opinions in the user-item graph and propose the framework GraphRec, which coherently models two graphs and heterogeneous strengths. Extensive experiments on two real-world datasets demonstrate the effectiveness of the proposed framework GraphRec. Our code is available at \url{https://github.com/wenqifan03/GraphRec-WWW19}
Tasks	Recommendation Systems
Published	2019-02-19
URL	https://arxiv.org/abs/1902.07243v2
PDF	https://arxiv.org/pdf/1902.07243v2.pdf
PWC	https://paperswithcode.com/paper/graph-neural-networks-for-social
Repo	https://github.com/wenqifan03/GraphRec-WWW19
Framework	pytorch

Neural Cages for Detail-Preserving 3D Deformations


Title	Neural Cages for Detail-Preserving 3D Deformations
Authors	Wang Yifan, Noam Aigerman, Vladimir G. Kim, Siddhartha Chaudhuri, Olga Sorkine-Hornung
Abstract	We propose a novel learnable representation for detail-preserving shape deformation. The goal of our method is to warp a source shape to match the general structure of a target shape, while preserving the surface details of the source. Our method extends a traditional cage-based deformation technique, where the source shape is enclosed by a coarse control mesh termed \emph{cage}, and translations prescribed on the cage vertices are interpolated to any point on the source mesh via special weight functions. The use of this sparse cage scaffolding enables preserving surface details regardless of the shape’s intricacy and topology. Our key contribution is a novel neural network architecture for predicting deformations by controlling the cage. We incorporate a differentiable cage-based deformation module in our architecture, and train our network end-to-end. Our method can be trained with common collections of 3D models in an unsupervised fashion, without any cage-specific annotations. We demonstrate the utility of our method for synthesizing shape variations and deformation transfer.
Tasks
Published	2019-12-13
URL	https://arxiv.org/abs/1912.06395v2
PDF	https://arxiv.org/pdf/1912.06395v2.pdf
PWC	https://paperswithcode.com/paper/neural-cages-for-detail-preserving-3d
Repo	https://github.com/yifita/deep_cage
Framework	pytorch