October 18, 2019

2907 words 14 mins read

Paper Group ANR 505

Paper Group ANR 505

Deep Loopy Neural Network Model for Graph Structured Data Representation Learning. Beyond “How may I help you?": Assisting Customer Service Agents with Proactive Responses. JobComposer: Career Path Optimization via Multicriteria Utility Learning. Dynamics and Reachability of Learning Tasks. Learning Robust Representations for Automatic Target Recog …

Deep Loopy Neural Network Model for Graph Structured Data Representation Learning

Title Deep Loopy Neural Network Model for Graph Structured Data Representation Learning
Authors Jiawei Zhang
Abstract Existing deep learning models may encounter great challenges in handling graph structured data. In this paper, we introduce a new deep learning model for graph data specifically, namely the deep loopy neural network. Significantly different from the previous deep models, inside the deep loopy neural network, there exist a large number of loops created by the extensive connections among nodes in the input graph data, which makes model learning an infeasible task. To resolve such a problem, in this paper, we will introduce a new learning algorithm for the deep loopy neural network specifically. Instead of learning the model variables based on the original model, in the proposed learning algorithm, errors will be back-propagated through the edges in a group of extracted spanning trees. Extensive numerical experiments have been done on several real-world graph datasets, and the experimental results demonstrate the effectiveness of both the proposed model and the learning algorithm in handling graph data.
Tasks Representation Learning
Published 2018-05-19
URL https://arxiv.org/abs/1805.07504v2
PDF https://arxiv.org/pdf/1805.07504v2.pdf
PWC https://paperswithcode.com/paper/deep-loopy-neural-network-model-for-graph
Repo
Framework

Beyond “How may I help you?": Assisting Customer Service Agents with Proactive Responses

Title Beyond “How may I help you?": Assisting Customer Service Agents with Proactive Responses
Authors Mengting Wan, Xin Chen
Abstract We study the problem of providing recommended responses to customer service agents in live-chat dialogue systems. Smart-reply systems have been widely applied in real-world applications (e.g. Gmail, LinkedIn Messaging), where most of them can successfully recommend reactive responses. However, we observe a major limitation of current methods is that they generally have difficulties in suggesting proactive investigation act (e.g. “Do you perhaps have another account with us?") due to the lack of long-term context information, which indeed act as critical steps for customer service agents to collect information and resolve customers’ issues. Thus in this work, we propose an end-to-end method with special focus on suggesting proactive investigative questions to customer agents in Airbnb’s customer service live-chat system. Effectiveness of our proposed method can be validated through qualitative and quantitative results.
Tasks
Published 2018-11-26
URL http://arxiv.org/abs/1811.10686v1
PDF http://arxiv.org/pdf/1811.10686v1.pdf
PWC https://paperswithcode.com/paper/beyond-how-may-i-help-you-assisting-customer
Repo
Framework

JobComposer: Career Path Optimization via Multicriteria Utility Learning

Title JobComposer: Career Path Optimization via Multicriteria Utility Learning
Authors Richard J. Oentaryo, Xavier Jayaraj Siddarth Ashok, Ee-Peng Lim, Philips Kokoh Prasetyo
Abstract With online professional network platforms (OPNs, e.g., LinkedIn, Xing, etc.) becoming popular on the web, people are now turning to these platforms to create and share their professional profiles, to connect with others who share similar professional aspirations and to explore new career opportunities. These platforms however do not offer a long-term roadmap to guide career progression and improve workforce employability. The career trajectories of OPN users can serve as a reference but they are not always optimal. A career plan can also be devised through consultation with career coaches, whose knowledge may however be limited to a few industries. To address the above limitations, we present a novel data-driven approach dubbed JobComposer to automate career path planning and optimization. Its key premise is that the observed career trajectories in OPNs may not necessarily be optimal, and can be improved by learning to maximize the sum of payoffs attainable by following a career path. At its heart, JobComposer features a decomposition-based multicriteria utility learning procedure to achieve the best tradeoff among different payoff criteria in career path planning. Extensive studies using a city state-based OPN dataset demonstrate that JobComposer returns career paths better than other baseline methods and the actual career paths.
Tasks
Published 2018-09-04
URL http://arxiv.org/abs/1809.01062v1
PDF http://arxiv.org/pdf/1809.01062v1.pdf
PWC https://paperswithcode.com/paper/jobcomposer-career-path-optimization-via
Repo
Framework

Dynamics and Reachability of Learning Tasks

Title Dynamics and Reachability of Learning Tasks
Authors Alessandro Achille, Glen Mbeng, Stefano Soatto
Abstract We compute the transition probability between two learning tasks, and show that it decomposes into two factors. The first depends on the geometry of the loss landscape of a model trained on each task, independent of any particular model used. This is related to an information theoretic distance function, but is insufficient to predict success in transfer learning, as nearby tasks can be unreachable via fine-tuning. The second factor depends on the ease of traversing the path between two tasks. With this dynamic component, we derive strict lower bounds on the complexity necessary to learn a task starting from the solution to another, which is one of the most common forms of transfer learning.
Tasks Semantic Textual Similarity, Transfer Learning
Published 2018-10-04
URL https://arxiv.org/abs/1810.02440v2
PDF https://arxiv.org/pdf/1810.02440v2.pdf
PWC https://paperswithcode.com/paper/the-dynamics-of-differential-learning-i
Repo
Framework

Learning Robust Representations for Automatic Target Recognition

Title Learning Robust Representations for Automatic Target Recognition
Authors Justin A. Goodwin, Olivia M. Brown, Taylor W. Killian, Sung-Hyun Son
Abstract Radio frequency (RF) sensors are used alongside other sensing modalities to provide rich representations of the world. Given the high variability of complex-valued target responses, RF systems are susceptible to attacks masking true target characteristics from accurate identification. In this work, we evaluate different techniques for building robust classification architectures exploiting learned physical structure in received synthetic aperture radar signals of simulated 3D targets.
Tasks
Published 2018-11-26
URL http://arxiv.org/abs/1811.10714v1
PDF http://arxiv.org/pdf/1811.10714v1.pdf
PWC https://paperswithcode.com/paper/learning-robust-representations-for-automatic
Repo
Framework

Vicious Circle Principle and Logic Programs with Aggregates

Title Vicious Circle Principle and Logic Programs with Aggregates
Authors Michael Gelfond, Yuanlin Zhang
Abstract The paper presents a knowledge representation language $\mathcal{A}log$ which extends ASP with aggregates. The goal is to have a language based on simple syntax and clear intuitive and mathematical semantics. We give some properties of $\mathcal{A}log$, an algorithm for computing its answer sets, and comparison with other approaches.
Tasks
Published 2018-08-21
URL http://arxiv.org/abs/1808.07050v1
PDF http://arxiv.org/pdf/1808.07050v1.pdf
PWC https://paperswithcode.com/paper/vicious-circle-principle-and-logic-programs
Repo
Framework

Improved Learning of One-hidden-layer Convolutional Neural Networks with Overlaps

Title Improved Learning of One-hidden-layer Convolutional Neural Networks with Overlaps
Authors Simon S. Du, Surbhi Goel
Abstract We propose a new algorithm to learn a one-hidden-layer convolutional neural network where both the convolutional weights and the outputs weights are parameters to be learned. Our algorithm works for a general class of (potentially overlapping) patches, including commonly used structures for computer vision tasks. Our algorithm draws ideas from (1) isotonic regression for learning neural networks and (2) landscape analysis of non-convex matrix factorization problems. We believe these findings may inspire further development in designing provable algorithms for learning neural networks and other complex models.
Tasks
Published 2018-05-20
URL http://arxiv.org/abs/1805.07798v2
PDF http://arxiv.org/pdf/1805.07798v2.pdf
PWC https://paperswithcode.com/paper/improved-learning-of-one-hidden-layer
Repo
Framework

Higher-order Projected Power Iterations for Scalable Multi-Matching

Title Higher-order Projected Power Iterations for Scalable Multi-Matching
Authors Florian Bernard, Johan Thunberg, Paul Swoboda, Christian Theobalt
Abstract The matching of multiple objects (e.g. shapes or images) is a fundamental problem in vision and graphics. In order to robustly handle ambiguities, noise and repetitive patterns in challenging real-world settings, it is essential to take geometric consistency between points into account. Computationally, the multi-matching problem is difficult. It can be phrased as simultaneously solving multiple (NP-hard) quadratic assignment problems (QAPs) that are coupled via cycle-consistency constraints. The main limitations of existing multi-matching methods are that they either ignore geometric consistency and thus have limited robustness, or they are restricted to small-scale problems due to their (relatively) high computational cost. We address these shortcomings by introducing a Higher-order Projected Power Iteration method, which is (i) efficient and scales to tens of thousands of points, (ii) straightforward to implement, (iii) able to incorporate geometric consistency, (iv) guarantees cycle-consistent multi-matchings, and (iv) comes with theoretical convergence guarantees. Experimentally we show that our approach is superior to existing methods.
Tasks
Published 2018-11-26
URL http://arxiv.org/abs/1811.10541v2
PDF http://arxiv.org/pdf/1811.10541v2.pdf
PWC https://paperswithcode.com/paper/higher-order-projected-power-iterations-for
Repo
Framework

Study and Observation of the Variations of Accuracies for Handwritten Digits Recognition with Various Hidden Layers and Epochs using Neural Network Algorithm

Title Study and Observation of the Variations of Accuracies for Handwritten Digits Recognition with Various Hidden Layers and Epochs using Neural Network Algorithm
Authors Md. Abu Bakr Siddique, Mohammad Mahmudur Rahman Khan, Rezoana Bente Arif, Zahidun Ashrafi
Abstract In recent days, Artificial Neural Network (ANN) can be applied to a vast majority of fields including business, medicine, engineering, etc. The most popular areas where ANN is employed nowadays are pattern and sequence recognition, novelty detection, character recognition, regression analysis, speech recognition, image compression, stock market prediction, Electronic nose, security, loan applications, data processing, robotics, and control. The benefits associated with its broad applications leads to increasing popularity of ANN in the era of 21st Century. ANN confers many benefits such as organic learning, nonlinear data processing, fault tolerance, and self-repairing compared to other conventional approaches. The primary objective of this paper is to analyze the influence of the hidden layers of a neural network over the overall performance of the network. To demonstrate this influence, we applied neural network with different layers on the MNIST dataset. Also, another goal is to observe the variations of accuracies of ANN for different numbers of hidden layers and epochs and to compare and contrast among them.
Tasks Image Compression, Speech Recognition, Stock Market Prediction
Published 2018-09-17
URL http://arxiv.org/abs/1809.06188v3
PDF http://arxiv.org/pdf/1809.06188v3.pdf
PWC https://paperswithcode.com/paper/study-and-observation-of-the-variations-of
Repo
Framework

Representing Images in 200 Bytes: Compression via Triangulation

Title Representing Images in 200 Bytes: Compression via Triangulation
Authors David Marwood, Pascal Massimino, Michele Covell, Shumeet Baluja
Abstract A rapidly increasing portion of internet traffic is dominated by requests from mobile devices with limited and metered bandwidth constraints. To satisfy these requests, it has become standard practice for websites to transmit small and extremely compressed image previews as part of the initial page load process to improve responsiveness. Increasing thumbnail compression beyond the capabilities of existing codecs is therefore an active research direction. In this work, we concentrate on extreme compression rates, where the size of the image is typically 200 bytes or less. First, we propose a novel approach for image compression that, unlike commonly used methods, does not rely on block-based statistics. We use an approach based on an adaptive triangulation of the target image, devoting more triangles to high entropy regions of the image. Second, we present a novel algorithm for encoding the triangles. The results show favorable statistics, in terms of PSNR and SSIM, over both the JPEG and the WebP standards.
Tasks Image Compression
Published 2018-09-07
URL http://arxiv.org/abs/1809.02257v2
PDF http://arxiv.org/pdf/1809.02257v2.pdf
PWC https://paperswithcode.com/paper/representing-images-in-200-bytes-compression
Repo
Framework

Analyzing and learning the language for different types of harassment

Title Analyzing and learning the language for different types of harassment
Authors Mohammadreza Rezvan, Saeedeh Shekarpour, Faisal Alshargi, Krishnaprasad Thirunarayan, Valerie L. Shalin, Amit Sheth
Abstract Disclaimer: This paper is concerned with violent online harassment. To describe the subject at an adequate level of realism, examples of our collected tweets involve violent, threatening, vulgar and hateful speech language in the context of racial, sexual, political, appearance and intellectual harassment. The presence of a significant amount of harassment in user-generated content and its negative impact calls for robust automatic detection approaches. This requires that we can identify different forms or types of harassment. Earlier work has classified harassing language in terms of hurtfulness, abusiveness, sentiment, and profanity. However, to identify and understand harassment more accurately, it is essential to determine the context that represents the interrelated conditions in which they occur. In this paper, we introduce the notion of contextual type to harassment involving five categories: (i) sexual, (ii) racial, (iii) appearance-related, (iv) intellectual and (v) political. We utilize an annotated corpus from Twitter distinguishing these types of harassment. To study the context for each type that sheds light on the linguistic meaning, interpretation, and distribution, we conduct two lines of investigation: an extensive linguistic analysis, and a statistical distribution of unigrams. We then build type-ware classifiers to automate the identification of type-specific harassment. Our experiments demonstrate that these classifiers provide competitive accuracy for identifying and analyzing harassment on social media. We present extensive discussion and major observations about the effectiveness of type-aware classifiers using a detailed comparison setup providing insight into the role of type-dependent features.
Tasks
Published 2018-11-01
URL https://arxiv.org/abs/1811.00644v2
PDF https://arxiv.org/pdf/1811.00644v2.pdf
PWC https://paperswithcode.com/paper/analyzing-and-learning-the-language-for
Repo
Framework

ExGate: Externally Controlled Gating for Feature-based Attention in Artificial Neural Networks

Title ExGate: Externally Controlled Gating for Feature-based Attention in Artificial Neural Networks
Authors Jarryd Son, Amit Mishra
Abstract Perceptual capabilities of artificial systems have come a long way since the advent of deep learning. These methods have proven to be effective, however they are not as efficient as their biological counterparts. Visual attention is a set of mechanisms that are employed in biological visual systems to ease computational load by only processing pertinent parts of the stimuli. This paper addresses the implementation of top-down, feature-based attention in an artificial neural network by use of externally controlled neuron gating. Our results showed a 5% increase in classification accuracy on the CIFAR-10 dataset versus a non-gated version, while adding very few parameters. Our gated model also produces more reasonable errors in predictions by drastically reducing prediction of classes that belong to a different category to the true class.
Tasks
Published 2018-11-08
URL http://arxiv.org/abs/1811.03403v1
PDF http://arxiv.org/pdf/1811.03403v1.pdf
PWC https://paperswithcode.com/paper/exgate-externally-controlled-gating-for
Repo
Framework

Measuring Human Assessed Complexity in Synthetic Aperture Sonar Imagery Using the Elo Rating System

Title Measuring Human Assessed Complexity in Synthetic Aperture Sonar Imagery Using the Elo Rating System
Authors Brian Reinhardt, Isaac Gerg, Daniel Brown, Joonho Park
Abstract Performance of automatic target recognition from synthetic aperture sonar data is heavily dependent on the complexity of the beamformed imagery. Several mechanisms can contribute to this, including unwanted vehicle dynamics, the bathymetry of the scene, and the presence of natural and manmade clutter. To understand the impact of the environmental complexity on image perception, researchers have taken approaches rooted in information theory, or heuristics. Despite these efforts, a quantitative measure for complexity has not been related to the phenomenology from which it is derived. By using subject matter experts (SMEs) we derive a complexity metric for a set of imagery which accounts for the underlying phenomenology. The goal of this work is to develop an understanding of how several common information theoretic and heuristic measures are related to the SME perceived complexity in synthetic aperture sonar imagery. To achieve this, an ensemble of 10-meter x 10-meter images were cropped from a high-frequency SAS data set that spans multiple environments. The SME’s were presented pairs of images from which they could rate the relative image complexity. These comparisons were then converted into the desired sequential ranking using a method first developed by A. Elo for establishing rankings of chess players. The Elo method produced a plausible rank ordering across the broad dataset. The heuristic and information theoretical metrics were then compared to the image rank from which they were derived. The metrics with the highest degree of correlation were those relating to spatial information, e.g. variations in pixel intensity, with an R-squared value of approximately 0.9. However, this agreement was dependent on the scale from which the spatial variation was measured. Results will also be presented for many other measures including lacunarity, image compression, and entropy.
Tasks Image Compression
Published 2018-08-15
URL http://arxiv.org/abs/1808.05279v1
PDF http://arxiv.org/pdf/1808.05279v1.pdf
PWC https://paperswithcode.com/paper/measuring-human-assessed-complexity-in
Repo
Framework

Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules

Title Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules
Authors Boris Knyazev, Xiao Lin, Mohamed R. Amer, Graham W. Taylor
Abstract Spectral Graph Convolutional Networks (GCNs) are a generalization of convolutional networks to learning on graph-structured data. Applications of spectral GCNs have been successful, but limited to a few problems where the graph is fixed, such as shape correspondence and node classification. In this work, we address this limitation by revisiting a particular family of spectral graph networks, Chebyshev GCNs, showing its efficacy in solving graph classification tasks with a variable graph structure and size. Chebyshev GCNs restrict graphs to have at most one edge between any pair of nodes. To this end, we propose a novel multigraph network that learns from multi-relational graphs. We model learned edges with abstract meaning and experiment with different ways to fuse the representations extracted from annotated and learned edges, achieving competitive results on a variety of chemical classification benchmarks.
Tasks Graph Classification, Node Classification
Published 2018-11-23
URL http://arxiv.org/abs/1811.09595v1
PDF http://arxiv.org/pdf/1811.09595v1.pdf
PWC https://paperswithcode.com/paper/spectral-multigraph-networks-for-discovering
Repo
Framework
Title Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search
Authors Bichen Wu, Yanghan Wang, Peizhao Zhang, Yuandong Tian, Peter Vajda, Kurt Keutzer
Abstract Recent work in network quantization has substantially reduced the time and space complexity of neural network inference, enabling their deployment on embedded and mobile devices with limited computational and memory resources. However, existing quantization methods often represent all weights and activations with the same precision (bit-width). In this paper, we explore a new dimension of the design space: quantizing different layers with different bit-widths. We formulate this problem as a neural architecture search problem and propose a novel differentiable neural architecture search (DNAS) framework to efficiently explore its exponential search space with gradient-based optimization. Experiments show we surpass the state-of-the-art compression of ResNet on CIFAR-10 and ImageNet. Our quantized models with 21.1x smaller model size or 103.9x lower computational cost can still outperform baseline quantized or even full precision models.
Tasks Neural Architecture Search, Quantization
Published 2018-11-30
URL http://arxiv.org/abs/1812.00090v1
PDF http://arxiv.org/pdf/1812.00090v1.pdf
PWC https://paperswithcode.com/paper/mixed-precision-quantization-of-convnets-via
Repo
Framework
comments powered by Disqus