January 26, 2020

3432 words 17 mins read

Paper Group ANR 1540

A memory enhanced LSTM for modeling complex temporal dependencies. Adversarial Representation Learning for Text-to-Image Matching. Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability. Learning Discrepancy Models From Experimental Data. The Usage of Web Search for Software Engineering. How much do you p …

A memory enhanced LSTM for modeling complex temporal dependencies


Title	A memory enhanced LSTM for modeling complex temporal dependencies
Authors	Sneha Aenugu
Abstract	In this paper, we present Gamma-LSTM, an enhanced long short term memory (LSTM) unit, to enable learning of hierarchical representations through multiple stages of temporal abstractions. Gamma memory, a hierarchical memory unit, forms the central memory of Gamma-LSTM with gates to regulate the information flow into various levels of hierarchy, thus providing the unit with a control to pick the appropriate level of hierarchy to process the input at a given instant of time. We demonstrate better performance of Gamma-LSTM model regular and stacked LSTMs in two settings (pixel-by-pixel MNIST digit classification and natural language inference) placing emphasis on the ability to generalize over long sequences.
Tasks	Natural Language Inference
Published	2019-10-25
URL	https://arxiv.org/abs/1910.12388v1
PDF	https://arxiv.org/pdf/1910.12388v1.pdf
PWC	https://paperswithcode.com/paper/a-memory-enhanced-lstm-for-modeling-complex
Repo
Framework

Adversarial Representation Learning for Text-to-Image Matching


Title	Adversarial Representation Learning for Text-to-Image Matching
Authors	Nikolaos Sarafianos, Xiang Xu, Ioannis A. Kakadiaris
Abstract	For many computer vision applications such as image captioning, visual question answering, and person search, learning discriminative feature representations at both image and text level is an essential yet challenging problem. Its challenges originate from the large word variance in the text domain as well as the difficulty of accurately measuring the distance between the features of the two modalities. Most prior work focuses on the latter challenge, by introducing loss functions that help the network learn better feature representations but fail to account for the complexity of the textual input. With that in mind, we introduce TIMAM: a Text-Image Modality Adversarial Matching approach that learns modality-invariant feature representations using adversarial and cross-modal matching objectives. In addition, we demonstrate that BERT, a publicly-available language model that extracts word embeddings, can successfully be applied in the text-to-image matching domain. The proposed approach achieves state-of-the-art cross-modal matching performance on four widely-used publicly-available datasets resulting in absolute improvements ranging from 2% to 5% in terms of rank-1 accuracy.
Tasks	Image Captioning, Language Modelling, Person Search, Question Answering, Representation Learning, Visual Question Answering, Word Embeddings
Published	2019-08-28
URL	https://arxiv.org/abs/1908.10534v1
PDF	https://arxiv.org/pdf/1908.10534v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-representation-learning-for-text
Repo
Framework

Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability


Title	Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability
Authors	Valerio Bonometti, Charles Ringer, Mark Hall, Alex R. Wade, Anders Drachen
Abstract	Data-driven approaches which aim to identify and predict player engagement are becoming increasingly popular in games industry contexts. This is due to the growing practice of tracking and storing large volumes of in-game telemetries coupled with a desire to tailor the gaming experience to the end-user’s needs. These approaches are particularly useful not just for companies adopting Game-as-a-Service (GaaS) models (e.g. for re-engagement strategies) but also for those working under persistent content-delivery regimes (e.g. for better audience targeting). A major challenge for the latter is to build engagement models of the user which are data-efficient, holistic and can generalize across multiple game titles and genres with minimal adjustments. This work leverages a theoretical framework rooted in engagement and behavioural science research for building a model able to estimate engagement-related behaviours employing only a minimal set of game-agnostic metrics. Through a series of experiments we show how, by modelling early user-game interactions, this approach can make joint estimates of long-term survival time and churn probability across several single-player games in a range of genres. The model proposed is very suitable for industry applications since it relies on a minimal set of metrics and observations, scales well with the number of users and is explicitly designed to work across a diverse range of titles.
Tasks
Published	2019-05-27
URL	https://arxiv.org/abs/1905.10998v3
PDF	https://arxiv.org/pdf/1905.10998v3.pdf
PWC	https://paperswithcode.com/paper/modelling-early-user-game-interactions-for
Repo
Framework

Learning Discrepancy Models From Experimental Data


Title	Learning Discrepancy Models From Experimental Data
Authors	Kadierdan Kaheman, Eurika Kaiser, Benjamin Strom, J. Nathan Kutz, Steven L. Brunton
Abstract	First principles modeling of physical systems has led to significant technological advances across all branches of science. For nonlinear systems, however, small modeling errors can lead to significant deviations from the true, measured behavior. Even in mechanical systems, where the equations are assumed to be well-known, there are often model discrepancies corresponding to nonlinear friction, wind resistance, etc. Discovering models for these discrepancies remains an open challenge for many complex systems. In this work, we use the sparse identification of nonlinear dynamics (SINDy) algorithm to discover a model for the discrepancy between a simplified model and measurement data. In particular, we assume that the model mismatch can be sparsely represented in a library of candidate model terms. We demonstrate the efficacy of our approach on several examples including experimental data from a double pendulum on a cart. We further design and implement a feed-forward controller in simulations, showing improvement with a discrepancy model.
Tasks
Published	2019-09-18
URL	https://arxiv.org/abs/1909.08574v1
PDF	https://arxiv.org/pdf/1909.08574v1.pdf
PWC	https://paperswithcode.com/paper/learning-discrepancy-models-from-experimental
Repo
Framework

The Usage of Web Search for Software Engineering


Title	The Usage of Web Search for Software Engineering
Authors	Chetan Bansal, Thomas Zimmermann, Ahmed Hassan Awadallah, Nachiappan Nagappan
Abstract	Internet plays a key role in accomplishing many tasks. For many such tasks, web search is integral in finding relevant information. Similar to other domains, web search is also heavily used in software engineering (SE) to help with various SE specific tasks such as debugging, finding documentation, installation, etc. In this paper, we present the first large scale study on how web search is used in software engineering. We analyze the query logs from a major commercial web search engine. Being able to disambiguate software engineering queries from other queries is important for understanding the usage of web search in software engineering. We build a machine learning based classifier for distinguishing software engineering related search queries from other queries. We then define the taxonomy of intents behind the usage of web search by software engineers. This allows us to develop a better understanding of the various contexts in which web search is used in software engineering. We also analyze 1 million web search sessions to understand how software engineering related queries and sessions differ from other queries and sessions. Our results show that web search is heavily used for SE related search tasks. Finally, we discuss implications of this work on improving search engine support for developers and providing more effective and contextual assistance to developers using web resources. To our knowledge, this is the first study to fully characterize online search tasks in the software engineering context with a focus on query and session level differences.
Tasks
Published	2019-12-19
URL	https://arxiv.org/abs/1912.09519v1
PDF	https://arxiv.org/pdf/1912.09519v1.pdf
PWC	https://paperswithcode.com/paper/the-usage-of-web-search-for-software
Repo
Framework

How much do you perceive this? An analysis on perceptions of geometric features, personalities and emotions in virtual humans (Extended Version)


Title	How much do you perceive this? An analysis on perceptions of geometric features, personalities and emotions in virtual humans (Extended Version)
Authors	Victor Araujo, Rodolfo Migon Favaretto, Paulo Knob, Soraia Raupp Musse, Felipe Vilanova, Angelo Brandelli Costa
Abstract	This work aims to evaluate people’s perception regarding geometric features, personalities and emotions characteristics in virtual humans. For this, we use as a basis, a dataset containing the tracking files of pedestrians captured from spontaneous videos and visualized them as identical virtual humans. The goal is to focus on their behavior and not being distracted by other features. In addition to tracking files containing their positions, the dataset also contains pedestrian emotions and personalities detected using Computer Vision and Pattern Recognition techniques. We proceed with our analysis in order to answer the question if subjects can perceive geometric features as distances/speeds as well as emotions and personalities in video sequences when pedestrians are represented by virtual humans. Regarding the participants, an amount of 73 people volunteered for the experiment. The analysis was divided in two parts: i) evaluation on perception of geometric characteristics, such as density, angular variation, distances and speeds, and ii) evaluation on personality and emotion perceptions. Results indicate that, even without explaining to the participants the concepts of each personality or emotion and how they were calculated (considering geometric characteristics), in most of the cases, participants perceived the personality and emotion expressed by the virtual agents, in accordance with the available ground truth.
Tasks
Published	2019-04-24
URL	http://arxiv.org/abs/1904.11084v1
PDF	http://arxiv.org/pdf/1904.11084v1.pdf
PWC	https://paperswithcode.com/paper/how-much-do-you-perceive-this-an-analysis-on
Repo
Framework

Meemi: A Simple Method for Post-processing Cross-lingual Word Embeddings


Title	Meemi: A Simple Method for Post-processing Cross-lingual Word Embeddings
Authors	Yerai Doval, Jose Camacho-Collados, Luis Espinosa-Anke, Steven Schockaert
Abstract	Word embeddings have become a standard resource in the toolset of any Natural Language Processing practitioner. While monolingual word embeddings encode information about words in the context of a particular language, cross-lingual embeddings define a multilingual space where word embeddings from two or more languages are integrated together. Current state-of-the-art approaches learn these embeddings by aligning two disjoint monolingual vector spaces through an orthogonal transformation which preserves the structure of the monolingual counterparts. In this work, we propose to apply an additional transformation after this initial alignment step, which aims to bring the vector representations of a given word and its translations closer to their average. Since this additional transformation is non-orthogonal, it also affects the structure of the monolingual spaces. We show that our approach both improves the integration of the monolingual spaces as well as the quality of the monolingual spaces themselves. Furthermore, because our transformation can be applied to an arbitrary number of languages, we are able to effectively obtain a truly multilingual space. The resulting (monolingual and multilingual) spaces show consistent gains over the current state-of-the-art in standard intrinsic tasks, namely dictionary induction and word similarity, as well as in extrinsic tasks such as cross-lingual hypernym discovery and cross-lingual natural language inference.
Tasks	Cross-Lingual Natural Language Inference, Hypernym Discovery, Natural Language Inference, Word Embeddings
Published	2019-10-16
URL	https://arxiv.org/abs/1910.07221v3
PDF	https://arxiv.org/pdf/1910.07221v3.pdf
PWC	https://paperswithcode.com/paper/meemi-finding-the-middle-ground-in-cross
Repo
Framework

Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning


Title	Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning
Authors	Ying Wen, Yaodong Yang, Rui Luo, Jun Wang, Wei Pan
Abstract	Humans are capable of attributing latent mental contents such as beliefs or intentions to others. The social skill is critical in daily life for reasoning about the potential consequences of others’ behaviors so as to plan ahead. It is known that humans use such reasoning ability recursively by considering what others believe about their own beliefs. In this paper, we start from level-$1$ recursion and introduce a probabilistic recursive reasoning (PR2) framework for multi-agent reinforcement learning. Our hypothesis is that it is beneficial for each agent to account for how the opponents would react to its future behaviors. Under the PR2 framework, we adopt variational Bayes methods to approximate the opponents’ conditional policies, to which each agent finds the best response and then improve their own policies. We develop decentralized-training-decentralized-execution algorithms, namely PR2-Q and PR2-Actor-Critic, that are proved to converge in the self-play scenarios when there exists one Nash equilibrium. Our methods are tested on both the matrix game and the differential game, which have a non-trivial equilibrium where common gradient-based methods fail to converge. Our experiments show that it is critical to reason about how the opponents believe about what the agent believes. We expect our work to contribute a new idea of modeling the opponents to the multi-agent reinforcement learning community.
Tasks	Multi-agent Reinforcement Learning
Published	2019-01-26
URL	http://arxiv.org/abs/1901.09207v2
PDF	http://arxiv.org/pdf/1901.09207v2.pdf
PWC	https://paperswithcode.com/paper/probabilistic-recursive-reasoning-for-multi
Repo
Framework

Debiasing Embeddings for Reduced Gender Bias in Text Classification


Title	Debiasing Embeddings for Reduced Gender Bias in Text Classification
Authors	Flavien Prost, Nithum Thain, Tolga Bolukbasi
Abstract	(Bolukbasi et al., 2016) demonstrated that pretrained word embeddings can inherit gender bias from the data they were trained on. We investigate how this bias affects downstream classification tasks, using the case study of occupation classification (De-Arteaga et al.,2019). We show that traditional techniques for debiasing embeddings can actually worsen the bias of the downstream classifier by providing a less noisy channel for communicating gender information. With a relatively minor adjustment, however, we show how these same techniques can be used to simultaneously reduce bias and maintain high classification accuracy.
Tasks	Text Classification, Word Embeddings
Published	2019-08-07
URL	https://arxiv.org/abs/1908.02810v1
PDF	https://arxiv.org/pdf/1908.02810v1.pdf
PWC	https://paperswithcode.com/paper/debiasing-embeddings-for-reduced-gender-bias-1
Repo
Framework

FusionNet: Incorporating Shape and Texture for Abnormality Detection in 3D Abdominal CT Scans


Title	FusionNet: Incorporating Shape and Texture for Abnormality Detection in 3D Abdominal CT Scans
Authors	Fengze Liu, Yuyin Zhou, Elliot Fishman, Alan Yuille
Abstract	Automatic abnormality detection in abdominal CT scans can help doctors improve the accuracy and efficiency in diagnosis. In this paper we aim at detecting pancreatic ductal adenocarcinoma (PDAC), the most common pancreatic cancer. Taking the fact that the existence of tumor can affect both the shape and the texture of pancreas, we design a system to extract the shape and texture feature at the same time for detecting PDAC. In this paper we propose a two-stage method for this 3D classification task. First, we segment the pancreas into a binary mask. Second, a FusionNet is proposed to take both the binary mask and CT image as input and perform a binary classification. The optimal architecture of the FusionNet is obtained by searching a pre-defined functional space. We show that the classification results using either shape or texture information are complementary, and by fusing them with the optimized architecture, the performance improves by a large margin. Our method achieves a specificity of 97% and a sensitivity of 92% on 200 normal scans and 136 scans with PDAC.
Tasks	Anomaly Detection
Published	2019-08-21
URL	https://arxiv.org/abs/1908.07654v2
PDF	https://arxiv.org/pdf/1908.07654v2.pdf
PWC	https://paperswithcode.com/paper/190807654
Repo
Framework

On the Effects of Pseudo and Quantum Random Number Generators in Soft Computing


Title	On the Effects of Pseudo and Quantum Random Number Generators in Soft Computing
Authors	Jordan J. Bird, Anikó Ekárt, Diego R. Faria
Abstract	In this work, we argue that the implications of Pseudo and Quantum Random Number Generators (PRNG and QRNG) inexplicably affect the performances and behaviours of various machine learning models that require a random input. These implications are yet to be explored in Soft Computing until this work. We use a CPU and a QPU to generate random numbers for multiple Machine Learning techniques. Random numbers are employed in the random initial weight distributions of Dense and Convolutional Neural Networks, in which results show a profound difference in learning patterns for the two. In 50 Dense Neural Networks (25 PRNG/25 QRNG), QRNG increases over PRNG for accent classification at +0.1%, and QRNG exceeded PRNG for mental state EEG classification by +2.82%. In 50 Convolutional Neural Networks (25 PRNG/25 QRNG), the MNIST and CIFAR-10 problems are benchmarked, in MNIST the QRNG experiences a higher starting accuracy than the PRNG but ultimately only exceeds it by 0.02%. In CIFAR-10, the QRNG outperforms PRNG by +0.92%. The n-random split of a Random Tree is enhanced towards and new Quantum Random Tree (QRT) model, which has differing classification abilities to its classical counterpart, 200 trees are trained and compared (100 PRNG/100 QRNG). Using the accent and EEG classification datasets, a QRT seemed inferior to a RT as it performed on average worse by -0.12%. This pattern is also seen in the EEG classification problem, where a QRT performs worse than a RT by -0.28%. Finally, the QRT is ensembled into a Quantum Random Forest (QRF), which also has a noticeable effect when compared to the standard Random Forest (RF)… ABSTRACT SHORTENED DUE TO ARXIV LIMIT
Tasks	EEG
Published	2019-10-10
URL	https://arxiv.org/abs/1910.04701v1
PDF	https://arxiv.org/pdf/1910.04701v1.pdf
PWC	https://paperswithcode.com/paper/on-the-effects-of-pseudo-and-quantum-random
Repo
Framework

Adversarial Examples in RF Deep Learning: Detection of the Attack and its Physical Robustness


Title	Adversarial Examples in RF Deep Learning: Detection of the Attack and its Physical Robustness
Authors	Silvija Kokalj-Filipovic, Rob Miller
Abstract	While research on adversarial examples in machine learning for images has been prolific, similar attacks on deep learning (DL) for radio frequency (RF) signals and their mitigation strategies are scarcely addressed in the published work, with only one recent publication in the RF domain [1]. RF adversarial examples (AdExs) can cause drastic, targeted misclassification results mostly in spectrum sensing/ survey applications (e.g. BPSK mistaken for 8-PSK) with minimal waveform perturbation. It is not clear if the RF AdExs maintain their effects in the physical world, i.e., when AdExs are delivered over-the-air (OTA). Our research on deep learning AdExs and proposed defense mechanisms are RF-centric, and incorporate physical world, OTA effects. We here present defense mechanisms based on statistical tests. One test to detect AdExs utilizes Peak-to- Average-Power-Ratio (PAPR) of the DL data points delivered OTA, while another statistical test uses the Softmax outputs of the DL classifier, which corresponds to the probabilities the classifier assigns to each of the trained classes. The former test leverages the RF nature of the data, and the latter is universally applicable to AdExs regardless of their origin. Both solutions are shown as viable mitigation methods to subvert adversarial attacks against communications and radar sensing systems.
Tasks
Published	2019-02-16
URL	http://arxiv.org/abs/1902.06044v1
PDF	http://arxiv.org/pdf/1902.06044v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-examples-in-rf-deep-learning
Repo
Framework

Online tuning and light source control using a physics-informed Gaussian process Adi


Title	Online tuning and light source control using a physics-informed Gaussian process Adi
Authors	A. Hanuka, J. Duris, J. Shtalenkova, D. Kennedy, A. Edelen, D. Ratner, X. Huang
Abstract	Operating large-scale scientific facilities often requires fast tuning and robust control in a high dimensional space. In this paper we introduce a new physics-informed optimization algorithm based on Gaussian process regression. Our method takes advantage of the existing domain knowledge in the form of realizations of a physics model of the observed system. We have applied a physics-informed Gaussian Process method experimentally at the SPEAR3 storage ring to demonstrate online accelerator optimization. This method outperforms Gaussian Process trained on data as well as the standard approach routinely used for operation, in terms of convergence speed and optimal point. The proposed method could be applicable to automatic tuning and control of other complex systems, without a prerequisite for any observed data.
Tasks
Published	2019-11-04
URL	https://arxiv.org/abs/1911.01538v1
PDF	https://arxiv.org/pdf/1911.01538v1.pdf
PWC	https://paperswithcode.com/paper/online-tuning-and-light-source-control-using
Repo
Framework

GCN-GAN: A Non-linear Temporal Link Prediction Model for Weighted Dynamic Networks


Title	GCN-GAN: A Non-linear Temporal Link Prediction Model for Weighted Dynamic Networks
Authors	Kai Lei, Meng Qin, Bo Bai, Gong Zhang, Min Yang
Abstract	In this paper, we generally formulate the dynamics prediction problem of various network systems (e.g., the prediction of mobility, traffic and topology) as the temporal link prediction task. Different from conventional techniques of temporal link prediction that ignore the potential non-linear characteristics and the informative link weights in the dynamic network, we introduce a novel non-linear model GCN-GAN to tackle the challenging temporal link prediction task of weighted dynamic networks. The proposed model leverages the benefits of the graph convolutional network (GCN), long short-term memory (LSTM) as well as the generative adversarial network (GAN). Thus, the dynamics, topology structure and evolutionary patterns of weighted dynamic networks can be fully exploited to improve the temporal link prediction performance. Concretely, we first utilize GCN to explore the local topological characteristics of each single snapshot and then employ LSTM to characterize the evolving features of the dynamic networks. Moreover, GAN is used to enhance the ability of the model to generate the next weighted network snapshot, which can effectively tackle the sparsity and the wide-value-range problem of edge weights in real-life dynamic networks. To verify the model’s effectiveness, we conduct extensive experiments on four datasets of different network systems and application scenarios. The experimental results demonstrate that our model achieves impressive results compared to the state-of-the-art competitors.
Tasks	Link Prediction
Published	2019-01-26
URL	http://arxiv.org/abs/1901.09165v1
PDF	http://arxiv.org/pdf/1901.09165v1.pdf
PWC	https://paperswithcode.com/paper/gcn-gan-a-non-linear-temporal-link-prediction
Repo
Framework

Unsupervised Polyglot Text To Speech


Title	Unsupervised Polyglot Text To Speech
Authors	Eliya Nachmani, Lior Wolf
Abstract	We present a TTS neural network that is able to produce speech in multiple languages. The proposed network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages. Training is done without using matching or parallel data, i.e., without samples of the same speaker in multiple languages, making the method much more applicable. The conversion is based on learning a polyglot network that has multiple per-language sub-networks and adding loss terms that preserve the speaker’s identity in multiple languages. We evaluate the proposed polyglot neural network for three languages with a total of more than 400 speakers and demonstrate convincing conversion capabilities.
Tasks
Published	2019-02-06
URL	http://arxiv.org/abs/1902.02263v1
PDF	http://arxiv.org/pdf/1902.02263v1.pdf
PWC	https://paperswithcode.com/paper/unsupervised-polyglot-text-to-speech
Repo
Framework