April 2, 2020

3381 words 16 mins read

Paper Group ANR 367

How To Avoid Being Eaten By a Grue: Exploration Strategies for Text-Adventure Agents. Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions. Multi-Feature Discrete Collaborative Filtering for Fast Cold-start Recommendation. A novel initialisation based on hospital-resident assignment for the k-modes algorithm. Semantic, Ef …

How To Avoid Being Eaten By a Grue: Exploration Strategies for Text-Adventure Agents


Title	How To Avoid Being Eaten By a Grue: Exploration Strategies for Text-Adventure Agents
Authors	Prithviraj Ammanabrolu, Ethan Tien, Zhaochen Luo, Mark O. Riedl
Abstract	Text-based games – in which an agent interacts with the world through textual natural language – present us with the problem of combinatorially-sized action-spaces. Most current reinforcement learning algorithms are not capable of effectively handling such a large number of possible actions per turn. Poor sample efficiency, consequently, results in agents that are unable to pass bottleneck states, where they are unable to proceed because they do not see the right action sequence to pass the bottleneck enough times to be sufficiently reinforced. Building on prior work using knowledge graphs in reinforcement learning, we introduce two new game state exploration strategies. We compare our exploration strategies against strong baselines on the classic text-adventure game, Zork1, where prior agent have been unable to get past a bottleneck where the agent is eaten by a Grue.
Tasks	Knowledge Graphs
Published	2020-02-19
URL	https://arxiv.org/abs/2002.08795v1
PDF	https://arxiv.org/pdf/2002.08795v1.pdf
PWC	https://paperswithcode.com/paper/how-to-avoid-being-eaten-by-a-grue
Repo
Framework

Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions


Title	Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions
Authors	Kamran Chitsaz, Mohsen Hajabdollahi, Nader Karimi, Shadrokh Samavi, Shahram Shirani
Abstract	Convolutional neural networks (CNNs) have a large number of variables and hence suffer from a complexity problem for their implementation. Different methods and techniques have developed to alleviate the problem of CNN’s complexity, such as quantization, pruning, etc. Among the different simplification methods, computation in the Fourier domain is regarded as a new paradigm for the acceleration of CNNs. Recent studies on Fast Fourier Transform (FFT) based CNN aiming at simplifying the computations required for FFT. However, there is a lot of space for working on the reduction of the computational complexity of FFT. In this paper, a new method for CNN processing in the FFT domain is proposed, which is based on input splitting. There are problems in the computation of FFT using small kernels in situations such as CNN. Splitting can be considered as an effective solution for such issues aroused by small kernels. Using splitting redundancy, such as overlap-and-add, is reduced and, efficiency is increased. Hardware implementation of the proposed FFT method, as well as different analyses of the complexity, are performed to demonstrate the proper performance of the proposed method.
Tasks	Quantization
Published	2020-03-27
URL	https://arxiv.org/abs/2003.12621v1
PDF	https://arxiv.org/pdf/2003.12621v1.pdf
PWC	https://paperswithcode.com/paper/acceleration-of-convolutional-neural-network
Repo
Framework

Multi-Feature Discrete Collaborative Filtering for Fast Cold-start Recommendation


Title	Multi-Feature Discrete Collaborative Filtering for Fast Cold-start Recommendation
Authors	Yang Xu, Lei Zhu, Zhiyong Cheng, Jingjing Li, Jiande Sun
Abstract	Hashing is an effective technique to address the large-scale recommendation problem, due to its high computation and storage efficiency on calculating the user preferences on items. However, existing hashing-based recommendation methods still suffer from two important problems: 1) Their recommendation process mainly relies on the user-item interactions and single specific content feature. When the interaction history or the content feature is unavailable (the cold-start problem), their performance will be seriously deteriorated. 2) Existing methods learn the hash codes with relaxed optimization or adopt discrete coordinate descent to directly solve binary hash codes, which results in significant quantization loss or consumes considerable computation time. In this paper, we propose a fast cold-start recommendation method, called Multi-Feature Discrete Collaborative Filtering (MFDCF), to solve these problems. Specifically, a low-rank self-weighted multi-feature fusion module is designed to adaptively project the multiple content features into binary yet informative hash codes by fully exploiting their complementarity. Additionally, we develop a fast discrete optimization algorithm to directly compute the binary hash codes with simple operations. Experiments on two public recommendation datasets demonstrate that MFDCF outperforms the state-of-the-arts on various aspects.
Tasks	Quantization
Published	2020-03-24
URL	https://arxiv.org/abs/2003.10719v1
PDF	https://arxiv.org/pdf/2003.10719v1.pdf
PWC	https://paperswithcode.com/paper/multi-feature-discrete-collaborative
Repo
Framework

A novel initialisation based on hospital-resident assignment for the k-modes algorithm


Title	A novel initialisation based on hospital-resident assignment for the k-modes algorithm
Authors	Henry Wilde, Vincent Knight, Jonathan Gillard
Abstract	This paper presents a new way of selecting an initial solution for the k-modes algorithm that allows for a notion of mathematical fairness and a leverage of the data that the common initialisations from literature do not. The method, which utilises the Hospital-Resident Assignment Problem to find the set of initial cluster centroids, is compared with the current initialisations on both benchmark datasets and a body of newly generated artificial datasets. Based on this analysis, the proposed method is shown to outperform the other initialisations in the majority of cases, especially when the number of clusters is optimised. In addition, we find that our method outperforms the leading established method specifically for low-density data.
Tasks
Published	2020-02-07
URL	https://arxiv.org/abs/2002.02701v1
PDF	https://arxiv.org/pdf/2002.02701v1.pdf
PWC	https://paperswithcode.com/paper/a-novel-initialisation-based-on-hospital
Repo
Framework

Semantic, Efficient, and Secure Search over Encrypted Cloud Data


Title	Semantic, Efficient, and Secure Search over Encrypted Cloud Data
Authors	Fateh Boucenna
Abstract	Companies and individuals demand more and more storage space and computing power. For this purpose, several new technologies have been designed and implemented, such as the cloud computing. This technology provides its users with storage space and computing power according to their needs in a flexible and personalized way. However, the outsourced data such as emails, electronic health records, and company reports are sensitive and confidential. Therefore, It is primordial to protect the outsourced data against possible external attacks and the cloud server itself. That is why it is highly recommended to encrypt the sensitive data before being outsourced to a remote server. To perform searches over outsourced data, it is no longer possible to exploit traditional search engines given that these data are encrypted. Consequently, lots of searchable encryption (SE) schemes have been proposed in the literature. Three major research axes of searchable encryption area have been studied in the literature. The first axis consists in ensuring the security of the search approach. Indeed, the search process should be performed without decryption any data and without causing any sensitive information leakage. The second axis consists in studying the search performance. In fact, the encrypted indexes are less efficient than the plaintext indexes, which makes the searchable encryption schemes very slow in practice. More the approach is secure, less it is efficient, thus, the challenge consists in finding the best compromise between security and performance. Finally, the third research axis consists in the quality of the returned results in terms of relevance and recall. The problem is that the encryption of the index causes the degradation of the recall and the precision. Therefore, the goal is to propose a technique that is able to obtain almost the same result obtained in the traditional search.
Tasks
Published	2020-02-24
URL	https://arxiv.org/abs/2002.10294v1
PDF	https://arxiv.org/pdf/2002.10294v1.pdf
PWC	https://paperswithcode.com/paper/semantic-efficient-and-secure-search-over
Repo
Framework

Evolutionary Multi-Objective Optimization Framework for Mining Association Rules


Title	Evolutionary Multi-Objective Optimization Framework for Mining Association Rules
Authors	Shaik Tanveer Ul Huq, Vadlamani Ravi
Abstract	In this paper, two multi-objective optimization frameworks in two variants (i.e., NSGA-III-ARM-V1, NSGA-III-ARM-V2; and MOEAD-ARM-V1, MOEAD-ARM-V2) are proposed to find association rules from transactional datasets. The first framework uses Non-dominated sorting genetic algorithm III (NSGA-III) and the second uses Decomposition based multi-objective evolutionary algorithm (MOEA/D) to find the association rules which are diverse, non-redundant and non-dominated (having high objective function values). In both these frameworks, there is no need to specify minimum support and minimum confidence. In the first variant, support, confidence, and lift are considered as objective functions while in second, confidence, lift, and interestingness are considered as objective functions. These frameworks are tested on seven different kinds of datasets including two real-life bank datasets. Our study suggests that NSGA-III-ARM framework works better than MOEAD-ARM framework in both the variants across majority of the datasets.
Tasks
Published	2020-03-20
URL	https://arxiv.org/abs/2003.09158v1
PDF	https://arxiv.org/pdf/2003.09158v1.pdf
PWC	https://paperswithcode.com/paper/evolutionary-multi-objective-optimization-2
Repo
Framework

Finding novelty with uncertainty


Title	Finding novelty with uncertainty
Authors	Jacob C. Reinhold, Yufan He, Shizhong Han, Yunqiang Chen, Dashan Gao, Junghoon Lee, Jerry L. Prince, Aaron Carass
Abstract	Medical images are often used to detect and characterize pathology and disease; however, automatically identifying and segmenting pathology in medical images is challenging because the appearance of pathology across diseases varies widely. To address this challenge, we propose a Bayesian deep learning method that learns to translate healthy computed tomography images to magnetic resonance images and simultaneously calculates voxel-wise uncertainty. Since high uncertainty occurs in pathological regions of the image, this uncertainty can be used for unsupervised anomaly segmentation. We show encouraging experimental results on an unsupervised anomaly segmentation task by combining two types of uncertainty into a novel quantity we call scibilic uncertainty.
Tasks
Published	2020-02-11
URL	https://arxiv.org/abs/2002.04626v1
PDF	https://arxiv.org/pdf/2002.04626v1.pdf
PWC	https://paperswithcode.com/paper/finding-novelty-with-uncertainty
Repo
Framework

A Method for Estimating Reflectance map and Material using Deep Learning with Synthetic Dataset


Title	A Method for Estimating Reflectance map and Material using Deep Learning with Synthetic Dataset
Authors	Mingi Lim, Sung-eui Yoon
Abstract	The process of decomposing target images into their internal properties is a difficult task due to the inherent ill-posed nature of the problem. The lack of data required to train a network is a one of the reasons why the decomposing appearance task is difficult. In this paper, we propose a deep learning-based reflectance map prediction system for material estimation of target objects in the image, so as to alleviate the ill-posed problem that occurs in this image decomposition operation. We also propose a network architecture for Bidirectional Reflectance Distribution Function (BRDF) parameter estimation, environment map estimation. We also use synthetic data to solve the lack of data problems. We get out of the previously proposed Deep Learning-based network architecture for reflectance map, and we newly propose to use conditional Generative Adversarial Network (cGAN) structures for estimating the reflectance map, which enables better results in many applications. To improve the efficiency of learning in this structure, we newly utilized the loss function using the normal map of the target object.
Tasks
Published	2020-01-15
URL	https://arxiv.org/abs/2001.05372v1
PDF	https://arxiv.org/pdf/2001.05372v1.pdf
PWC	https://paperswithcode.com/paper/a-method-for-estimating-reflectance-map-and
Repo
Framework

Being Bayesian about Categorical Probability


Title	Being Bayesian about Categorical Probability
Authors	Taejong Joo, Uijung Chung, Min-Gwan Seo
Abstract	Neural networks utilize the softmax as a building block in classification tasks, which contains an overconfidence problem and lacks an uncertainty representation ability. As a Bayesian alternative to the softmax, we consider a random variable of a categorical probability over class labels. In this framework, the prior distribution explicitly models the presumed noise inherent in the observed label, which provides consistent gains in generalization performance in multiple challenging tasks. The proposed method inherits advantages of Bayesian approaches that achieve better uncertainty estimation and model calibration. Our method can be implemented as a plug-and-play loss function with negligible computational overhead compared to the softmax with the cross-entropy loss function.
Tasks	Calibration
Published	2020-02-19
URL	https://arxiv.org/abs/2002.07965v1
PDF	https://arxiv.org/pdf/2002.07965v1.pdf
PWC	https://paperswithcode.com/paper/being-bayesian-about-categorical-probability
Repo
Framework

On Thompson Sampling for Smoother-than-Lipschitz Bandits


Title	On Thompson Sampling for Smoother-than-Lipschitz Bandits
Authors	James A. Grant, David S. Leslie
Abstract	Thompson Sampling is a well established approach to bandit and reinforcement learning problems. However its use in continuum armed bandit problems has received relatively little attention. We provide the first bounds on the regret of Thompson Sampling for continuum armed bandits under weak conditions on the function class containing the true function and sub-exponential observation noise. Our bounds are realised by analysis of the eluder dimension, a recently proposed measure of the complexity of a function class, which has been demonstrated to be useful in bounding the Bayesian regret of Thompson Sampling for simpler bandit problems under sub-Gaussian observation noise. We derive a new bound on the eluder dimension for classes of functions with Lipschitz derivatives, and generalise previous analyses in multiple regards.
Tasks
Published	2020-01-08
URL	https://arxiv.org/abs/2001.02323v2
PDF	https://arxiv.org/pdf/2001.02323v2.pdf
PWC	https://paperswithcode.com/paper/on-thompson-sampling-for-smoother-than
Repo
Framework

Fact-aware Sentence Split and Rephrase with Permutation Invariant Training


Title	Fact-aware Sentence Split and Rephrase with Permutation Invariant Training
Authors	Yinuo Guo, Tao Ge, Furu Wei
Abstract	Sentence Split and Rephrase aims to break down a complex sentence into several simple sentences with its meaning preserved. Previous studies tend to address the issue by seq2seq learning from parallel sentence pairs, which takes a complex sentence as input and sequentially generates a series of simple sentences. However, the conventional seq2seq learning has two limitations for this task: (1) it does not take into account the facts stated in the long sentence; As a result, the generated simple sentences may miss or inaccurately state the facts in the original sentence. (2) The order variance of the simple sentences to be generated may confuse the seq2seq model during training because the simple sentences derived from the long source sentence could be in any order. To overcome the challenges, we first propose the Fact-aware Sentence Encoding, which enables the model to learn facts from the long sentence and thus improves the precision of sentence split; then we introduce Permutation Invariant Training to alleviate the effects of order variance in seq2seq learning for this task. Experiments on the WebSplit-v1.0 benchmark dataset show that our approaches can largely improve the performance over the previous seq2seq learning approaches. Moreover, an extrinsic evaluation on oie-benchmark verifies the effectiveness of our approaches by an observation that splitting long sentences with our state-of-the-art model as preprocessing is helpful for improving OpenIE performance.
Tasks
Published	2020-01-16
URL	https://arxiv.org/abs/2001.11383v2
PDF	https://arxiv.org/pdf/2001.11383v2.pdf
PWC	https://paperswithcode.com/paper/fact-aware-sentence-split-and-rephrase-with
Repo
Framework

Understanding the impact of mistakes on background regions in crowd counting


Title	Understanding the impact of mistakes on background regions in crowd counting
Authors	Davide Modolo, Bing Shuai, Rahul Rama Varior, Joseph Tighe
Abstract	Every crowd counting researcher has likely observed their model output wrong positive predictions on image regions not containing any person. But how often do these mistakes happen? Are our models negatively affected by this? In this paper we analyze this problem in depth. In order to understand its magnitude, we present an extensive analysis on five of the most important crowd counting datasets. We present this analysis in two parts. First, we quantify the number of mistakes made by popular crowd counting approaches. Our results show that (i) mistakes on background are substantial and they are responsible for 18-49% of the total error, (ii) models do not generalize well to different kinds of backgrounds and perform poorly on completely background images, and (iii) models make many more mistakes than those captured by the standard Mean Absolute Error (MAE) metric, as counting on background compensates considerably for misses on foreground. And second, we quantify the performance change gained by helping the model better deal with this problem. We enrich a typical crowd counting network with a segmentation branch trained to suppress background predictions. This simple addition (i) reduces background error by 10-83%, (ii) reduces foreground error by up to 26% and (iii) improves overall crowd counting performance up to 20%. When compared against the literature, this simple technique achieves very competitive results on all datasets, on par with the state-of-the-art, showing the importance of tackling the background problem.
Tasks	Crowd Counting
Published	2020-03-30
URL	https://arxiv.org/abs/2003.13759v1
PDF	https://arxiv.org/pdf/2003.13759v1.pdf
PWC	https://paperswithcode.com/paper/understanding-the-impact-of-mistakes-on
Repo
Framework

Evaluation of electrical efficiency of photovoltaic thermal solar collector


Title	Evaluation of electrical efficiency of photovoltaic thermal solar collector
Authors	Mohammad Hossein Ahmadi, Alireza Baghban, Milad Sadeghzadeh, Mohammad Zamen, Amir Mosavi, Shahaboddin Shamshirband, Ravinder Kumar, Mohammad Mohammadi-Khanaposhtani
Abstract	Solar energy is a renewable resource of energy that is broadly utilized and has the least emissions among renewable energies. In this study, machine learning methods of artificial neural networks (ANNs), least squares support vector machines (LSSVM), and neuro-fuzzy are used for advancing prediction models for the thermal performance of a photovoltaic-thermal solar collector (PV/T). In the proposed models, the inlet temperature, flow rate, heat, solar radiation, and the sun heat have been considered as the inputs variables. Data set has been extracted through experimental measurements from a novel solar collector system. Different analyses are performed to examine the credibility of the introduced approaches and evaluate their performance. The proposed LSSVM model outperformed ANFIS and ANNs models. LSSVM model is reported suitable when the laboratory measurements are costly and time-consuming, or achieving such values requires sophisticated interpretations.
Tasks
Published	2020-02-11
URL	https://arxiv.org/abs/2002.05542v1
PDF	https://arxiv.org/pdf/2002.05542v1.pdf
PWC	https://paperswithcode.com/paper/evaluation-of-electrical-efficiency-of
Repo
Framework

How to Find a Point in the Convex Hull Privately


Title	How to Find a Point in the Convex Hull Privately
Authors	Haim Kaplan, Micha Sharir, Uri Stemmer
Abstract	We study the question of how to compute a point in the convex hull of an input set $S$ of $n$ points in ${\mathbb R}^d$ in a differentially private manner. This question, which is trivial non-privately, turns out to be quite deep when imposing differential privacy. In particular, it is known that the input points must reside on a fixed finite subset $G\subseteq{\mathbb R}^d$, and furthermore, the size of $S$ must grow with the size of $G$. Previous works focused on understanding how $n$ needs to grow with $G$, and showed that $n=O\left(d^{2.5}\cdot8^{\log^*G}\right)$ suffices (so $n$ does not have to grow significantly with $G$). However, the available constructions exhibit running time at least $G^{d^2}$, where typically $G=X^d$ for some (large) discretization parameter $X$, so the running time is in fact $\Omega(X^{d^3})$. In this paper we give a differentially private algorithm that runs in $O(n^d)$ time, assuming that $n=\Omega(d^4\log X)$. To get this result we study and exploit some structural properties of the Tukey levels (the regions $D_{\ge k}$ consisting of points whose Tukey depth is at least $k$, for $k=0,1,…$). In particular, we derive lower bounds on their volumes for point sets $S$ in general position, and develop a rather subtle mechanism for handling point sets $S$ in degenerate position (where the deep Tukey regions have zero volume). A naive approach to the construction of the Tukey regions requires $n^{O(d^2)}$ time. To reduce the cost to $O(n^d)$, we use an approximation scheme for estimating the volumes of the Tukey regions (within their affine spans in case of degeneracy), and for sampling a point from such a region, a scheme that is based on the volume estimation framework of Lov'asz and Vempala (FOCS 2003) and of Cousins and Vempala (STOC 2015). Making this framework differentially private raises a set of technical challenges that we address.
Tasks
Published	2020-03-30
URL	https://arxiv.org/abs/2003.13192v1
PDF	https://arxiv.org/pdf/2003.13192v1.pdf
PWC	https://paperswithcode.com/paper/how-to-find-a-point-in-the-convex-hull
Repo
Framework

Clean-Label Backdoor Attacks on Video Recognition Models


Title	Clean-Label Backdoor Attacks on Video Recognition Models
Authors	Shihao Zhao, Xingjun Ma, Xiang Zheng, James Bailey, Jingjing Chen, Yu-Gang Jiang
Abstract	Deep neural networks (DNNs) are vulnerable to backdoor attacks which can hide backdoor triggers in DNNs by poisoning training data. A backdoored model behaves normally on clean test images, yet consistently predicts a particular target class for any test examples that contain the trigger pattern. As such, backdoor attacks are hard to detect, and have raised severe security concerns in real-world applications. Thus far, backdoor research has mostly been conducted in the image domain with image classification models. In this paper, we show that existing image backdoor attacks are far less effective on videos, and outline 4 strict conditions where existing attacks are likely to fail: 1) scenarios with more input dimensions (eg. videos), 2) scenarios with high resolution, 3) scenarios with a large number of classes and few examples per class (a “sparse dataset”), and 4) attacks with access to correct labels (eg. clean-label attacks). We propose the use of a universal adversarial trigger as the backdoor trigger to attack video recognition models, a situation where backdoor attacks are likely to be challenged by the above 4 strict conditions. We show on benchmark video datasets that our proposed backdoor attack can manipulate state-of-the-art video models with high success rates by poisoning only a small proportion of training data (without changing the labels). We also show that our proposed backdoor attack is resistant to state-of-the-art backdoor defense/detection methods, and can even be applied to improve image backdoor attacks. Our proposed video backdoor attack not only serves as a strong baseline for improving the robustness of video models, but also provides a new perspective for more understanding more powerful backdoor attacks.
Tasks	Image Classification, Video Recognition
Published	2020-03-06
URL	https://arxiv.org/abs/2003.03030v1
PDF	https://arxiv.org/pdf/2003.03030v1.pdf
PWC	https://paperswithcode.com/paper/clean-label-backdoor-attacks-on-video
Repo
Framework