Paper Group ANR 28
NeckSense: A Multi-Sensor Necklace for Detecting Eating Activities in Free-Living Conditions. Heavy-ball Algorithms Always Escape Saddle Points. Ambient Lighting Generation for Flash Images with Guided Conditional Adversarial Networks. Combating Adversarial Misspellings with Robust Word Recognition. Multi-Head Attention with Diversity for Learning …
NeckSense: A Multi-Sensor Necklace for Detecting Eating Activities in Free-Living Conditions
| Title | NeckSense: A Multi-Sensor Necklace for Detecting Eating Activities in Free-Living Conditions |
| Authors | Shibo Zhang, Yuqi Zhao, Dzung Tri Nguyen, Runsheng Xu, Sougata Sen, Josiah Hester, Nabil Alshurafa |
| Abstract | We present the design, implementation, and evaluation of a multi-sensor low-power necklace ‘NeckSense’ for automatically and unobtrusively capturing fine-grained information about an individual’s eating activity and eating episodes, across an entire waking-day in a naturalistic setting. The NeckSense fuses and classifies the proximity of the necklace from the chin, the ambient light, the Lean Forward Angle, and the energy signals to determine chewing sequences, a building block of the eating activity. It then clusters the identified chewing sequences to determine eating episodes. We tested NeckSense with 11 obese and 9 non-obese participants across two studies, where we collected more than 470 hours of data in naturalistic setting. Our result demonstrates that NeckSense enables reliable eating-detection for an entire waking-day, even in free-living environments. Overall, our system achieves an F1-score of 81.6% in detecting eating episodes in an exploratory study. Moreover, our system can achieve a F1-score of 77.1% for episodes even in an all-day-around free-living setting. With more than 15.8 hours of battery-life NeckSense will allow researchers and dietitians to better understand natural chewing and eating behaviors, and also enable real-time interventions. |
| Tasks | |
| Published | 2019-11-17 |
| URL | https://arxiv.org/abs/1911.07179v2 |
| https://arxiv.org/pdf/1911.07179v2.pdf | |
| PWC | https://paperswithcode.com/paper/necksense-a-multi-sensor-necklace-for |
| Repo | |
| Framework | |
Heavy-ball Algorithms Always Escape Saddle Points
| Title | Heavy-ball Algorithms Always Escape Saddle Points |
| Authors | Tao Sun, Dongsheng Li, Zhe Quan, Hao Jiang, Shengguo Li, Yong Dou |
| Abstract | Nonconvex optimization algorithms with random initialization have attracted increasing attention recently. It has been showed that many first-order methods always avoid saddle points with random starting points. In this paper, we answer a question: can the nonconvex heavy-ball algorithms with random initialization avoid saddle points? The answer is yes! Direct using the existing proof technique for the heavy-ball algorithms is hard due to that each iteration of the heavy-ball algorithm consists of current and last points. It is impossible to formulate the algorithms as iteration like xk+1= g(xk) under some mapping g. To this end, we design a new mapping on a new space. With some transfers, the heavy-ball algorithm can be interpreted as iterations after this mapping. Theoretically, we prove that heavy-ball gradient descent enjoys larger stepsize than the gradient descent to escape saddle points to escape the saddle point. And the heavy-ball proximal point algorithm is also considered; we also proved that the algorithm can always escape the saddle point. |
| Tasks | |
| Published | 2019-07-23 |
| URL | https://arxiv.org/abs/1907.09697v1 |
| https://arxiv.org/pdf/1907.09697v1.pdf | |
| PWC | https://paperswithcode.com/paper/heavy-ball-algorithms-always-escape-saddle |
| Repo | |
| Framework | |
Ambient Lighting Generation for Flash Images with Guided Conditional Adversarial Networks
| Title | Ambient Lighting Generation for Flash Images with Guided Conditional Adversarial Networks |
| Authors | José Chávez, Rensso Mora, Edward Cayllahua-Cahuina |
| Abstract | To cope with the challenges that low light conditions produce in images, photographers tend to use the light provided by the camera flash to get better illumination. Nevertheless, harsh shadows and non-uniform illumination can arise from using a camera flash, especially in low light conditions. Previous studies have focused on normalizing the lighting on flash images; however, to the best of our knowledge, no prior studies have examined the sideways shadows removal, reconstruction of overexposed areas, and the generation of synthetic ambient shadows or natural tone of scene objects. To provide more natural illumination on flash images and ensure high-frequency details, we propose a generative adversarial network in a guided conditional mode. We show that this approach not only generates natural illumination but also attenuates harsh shadows, simultaneously generating synthetic ambient shadows. Our approach achieves promising results on a custom FAID dataset, outperforming our baseline studies. We also analyze the components of our proposal and how they affect the overall performance and discuss the opportunities for future work. |
| Tasks | |
| Published | 2019-12-18 |
| URL | https://arxiv.org/abs/1912.08813v5 |
| https://arxiv.org/pdf/1912.08813v5.pdf | |
| PWC | https://paperswithcode.com/paper/ambient-lighting-generation-for-flash-images |
| Repo | |
| Framework | |
Combating Adversarial Misspellings with Robust Word Recognition
| Title | Combating Adversarial Misspellings with Robust Word Recognition |
| Authors | Danish Pruthi, Bhuwan Dhingra, Zachary C. Lipton |
| Abstract | To combat adversarial spelling mistakes, we propose placing a word recognition model in front of the downstream classifier. Our word recognition models build upon the RNN semi-character architecture, introducing several new backoff strategies for handling rare and unseen words. Trained to recognize words corrupted by random adds, drops, swaps, and keyboard mistakes, our method achieves 32% relative (and 3.3% absolute) error reduction over the vanilla semi-character model. Notably, our pipeline confers robustness on the downstream classifier, outperforming both adversarial training and off-the-shelf spell checkers. Against a BERT model fine-tuned for sentiment analysis, a single adversarially-chosen character attack lowers accuracy from 90.3% to 45.8%. Our defense restores accuracy to 75%. Surprisingly, better word recognition does not always entail greater robustness. Our analysis reveals that robustness also depends upon a quantity that we denote the sensitivity. |
| Tasks | Sentiment Analysis |
| Published | 2019-05-27 |
| URL | https://arxiv.org/abs/1905.11268v2 |
| https://arxiv.org/pdf/1905.11268v2.pdf | |
| PWC | https://paperswithcode.com/paper/combating-adversarial-misspellings-with |
| Repo | |
| Framework | |
Multi-Head Attention with Diversity for Learning Grounded Multilingual Multimodal Representations
| Title | Multi-Head Attention with Diversity for Learning Grounded Multilingual Multimodal Representations |
| Authors | Po-Yao Huang, Xiaojun Chang, Alexander Hauptmann |
| Abstract | With the aim of promoting and understanding the multilingual version of image search, we leverage visual object detection and propose a model with diverse multi-head attention to learn grounded multilingual multimodal representations. Specifically, our model attends to different types of textual semantics in two languages and visual objects for fine-grained alignments between sentences and images. We introduce a new objective function which explicitly encourages attention diversity to learn an improved visual-semantic embedding space. We evaluate our model in the German-Image and English-Image matching tasks on the Multi30K dataset, and in the Semantic Textual Similarity task with the English descriptions of visual content. Results show that our model yields a significant performance gain over other methods in all of the three tasks. |
| Tasks | Image Retrieval, Object Detection, Semantic Textual Similarity |
| Published | 2019-09-30 |
| URL | https://arxiv.org/abs/1910.00058v1 |
| https://arxiv.org/pdf/1910.00058v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-head-attention-with-diversity-for |
| Repo | |
| Framework | |
Automatic detection of estuarine dolphin whistles in spectrogram images
| Title | Automatic detection of estuarine dolphin whistles in spectrogram images |
| Authors | O. M. Serra, F. P. R. Martins, L. R. Padovese |
| Abstract | An algorithm for detecting tonal vocalizations from estuarine dolphin (Sotalia guianensis) specimens without interference of a human operator is developed. The raw audio data collected from a passive monitoring sensor in the Canan'eia underwater soundscape is converted to spectrogram images, containing the desired acoustic event (whistle) as a linear pattern in the images. Detection is a four-step method: first, ridge maps are obtained from the spectrogram images; second, a probabilistic Hough transform algorithm is applied to detect roughly linear ridges, which are adjusted to the true corresponding shape of the whistles via an active contour algorithm; third, feature vectors are built from the geometry of each detected curve; and fourth, the detections are fed to a random forest classifier to parse out false positives. We develop a system capable of reliably classifying roughly 97% of the characteristic patterns detected as Sotalia guianensis whistles or random empty detections. |
| Tasks | |
| Published | 2019-09-10 |
| URL | https://arxiv.org/abs/1909.04425v1 |
| https://arxiv.org/pdf/1909.04425v1.pdf | |
| PWC | https://paperswithcode.com/paper/automatic-detection-of-estuarine-dolphin |
| Repo | |
| Framework | |
Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation
| Title | Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation |
| Authors | Kyung-Min Kim, Donghyun Kwak, Hanock Kwak, Young-Jin Park, Sangkwon Sim, Jae-Han Cho, Minkyu Kim, Jihun Kwon, Nako Sung, Jung-Woo Ha |
| Abstract | Graph Neural Networks (GNNs) have been emerging as a promising method for relational representation including recommender systems. However, various challenging issues of social graphs hinder the practical usage of GNNs for social recommendation, such as their complex noisy connections and high heterogeneity. The oversmoothing of GNNs is an obstacle of GNN-based social recommendation as well. Here we propose a new graph embedding method Heterogeneous Graph Propagation (HGP) to tackle these issues. HGP uses a group-user-item tripartite graph as input to reduce the number of edges and the complexity of paths in a social graph. To solve the oversmoothing issue, HGP embeds nodes under a personalized PageRank based propagation scheme, separately for group-user graph and user-item graph. Node embeddings from each graph are integrated using an attention mechanism. We evaluate our HGP on a large-scale real-world dataset consisting of 1,645,279 nodes and 4,711,208 edges. The experimental results show that HGP outperforms several baselines in terms of AUC and F1-score metrics. |
| Tasks | Graph Embedding, Recommendation Systems |
| Published | 2019-07-24 |
| URL | https://arxiv.org/abs/1908.02569v1 |
| https://arxiv.org/pdf/1908.02569v1.pdf | |
| PWC | https://paperswithcode.com/paper/tripartite-heterogeneous-graph-propagation |
| Repo | |
| Framework | |
Citation Needed: A Taxonomy and Algorithmic Assessment of Wikipedia’s Verifiability
| Title | Citation Needed: A Taxonomy and Algorithmic Assessment of Wikipedia’s Verifiability |
| Authors | Miriam Redi, Besnik Fetahu, Jonathan Morgan, Dario Taraborelli |
| Abstract | Wikipedia is playing an increasingly central role on the web,and the policies its contributors follow when sourcing and fact-checking content affect million of readers. Among these core guiding principles, verifiability policies have a particularly important role. Verifiability requires that information included in a Wikipedia article be corroborated against reliable secondary sources. Because of the manual labor needed to curate and fact-check Wikipedia at scale, however, its contents do not always evenly comply with these policies. Citations (i.e. reference to external sources) may not conform to verifiability requirements or may be missing altogether, potentially weakening the reliability of specific topic areas of the free encyclopedia. In this paper, we aim to provide an empirical characterization of the reasons why and how Wikipedia cites external sources to comply with its own verifiability guidelines. First, we construct a taxonomy of reasons why inline citations are required by collecting labeled data from editors of multiple Wikipedia language editions. We then collect a large-scale crowdsourced dataset of Wikipedia sentences annotated with categories derived from this taxonomy. Finally, we design and evaluate algorithmic models to determine if a statement requires a citation, and to predict the citation reason based on our taxonomy. We evaluate the robustness of such models across different classes of Wikipedia articles of varying quality, as well as on an additional dataset of claims annotated for fact-checking purposes. |
| Tasks | |
| Published | 2019-02-28 |
| URL | http://arxiv.org/abs/1902.11116v1 |
| http://arxiv.org/pdf/1902.11116v1.pdf | |
| PWC | https://paperswithcode.com/paper/citation-needed-a-taxonomy-and-algorithmic |
| Repo | |
| Framework | |
Investigation of wind pressures on tall building under interference effects using machine learning techniques
| Title | Investigation of wind pressures on tall building under interference effects using machine learning techniques |
| Authors | Gang Hu, Lingbo Liu, Dacheng Tao, Jie Song, K. C. S. Kwok |
| Abstract | Interference effects of tall buildings have attracted numerous studies due to the boom of clusters of tall buildings in megacities. To fully understand the interference effects of buildings, it often requires a substantial amount of wind tunnel tests. Limited wind tunnel tests that only cover part of interference scenarios are unable to fully reveal the interference effects. This study used machine learning techniques to resolve the conflicting requirement between limited wind tunnel tests that produce unreliable results and a completed investigation of the interference effects that is costly and time-consuming. Four machine learning models including decision tree, random forest, XGBoost, generative adversarial networks (GANs), were trained based on 30% of a dataset to predict both mean and fluctuating pressure coefficients on the principal building. The GANs model exhibited the best performance in predicting these pressure coefficients. A number of GANs models were then trained based on different portions of the dataset ranging from 10% to 90%. It was found that the GANs model based on 30% of the dataset is capable of predicting both mean and fluctuating pressure coefficients under unseen interference conditions accurately. By using this GANs model, 70% of the wind tunnel test cases can be saved, largely alleviating the cost of this kind of wind tunnel testing study. |
| Tasks | |
| Published | 2019-08-20 |
| URL | https://arxiv.org/abs/1908.07307v1 |
| https://arxiv.org/pdf/1908.07307v1.pdf | |
| PWC | https://paperswithcode.com/paper/investigation-of-wind-pressures-on-tall |
| Repo | |
| Framework | |
Using Google Analytics to Support Cybersecurity Forensics
| Title | Using Google Analytics to Support Cybersecurity Forensics |
| Authors | Han Qin, Kit Riehle, Haozhen Zhao |
| Abstract | Web traffic is a valuable data source, typically used in the marketing space to track brand awareness and advertising effectiveness. However, web traffic is also a rich source of information for cybersecurity monitoring efforts. To better understand the threat of malicious cyber actors, this study develops a methodology to monitor and evaluate web activity using data archived from Google Analytics. Google Analytics collects and aggregates web traffic, including information about web visitors’ location, date and time of visit, visited webpages, and searched keywords. This study seeks to streamline analysis of this data and uses rule-based anomaly detection and predictive modeling to identify web traffic that deviates from normal patterns. Rather than evaluating pieces of web traffic individually, the methodology seeks to emulate real user behavior by creating a new unit of analysis: the user session. User sessions group individual pieces of traffic from the same location and date, which transforms the available information from single point-in-time snapshots to dynamic sessions showing users’ trajectory and intent. The result is faster and better insight into large volumes of noisy web traffic. |
| Tasks | Anomaly Detection |
| Published | 2019-04-03 |
| URL | http://arxiv.org/abs/1904.01725v1 |
| http://arxiv.org/pdf/1904.01725v1.pdf | |
| PWC | https://paperswithcode.com/paper/using-google-analytics-to-support |
| Repo | |
| Framework | |
Error bounds for approximations with deep ReLU neural networks in $W^{s,p}$ norms
| Title | Error bounds for approximations with deep ReLU neural networks in $W^{s,p}$ norms |
| Authors | Ingo Gühring, Gitta Kutyniok, Philipp Petersen |
| Abstract | We analyze approximation rates of deep ReLU neural networks for Sobolev-regular functions with respect to weaker Sobolev norms. First, we construct, based on a calculus of ReLU networks, artificial neural networks with ReLU activation functions that achieve certain approximation rates. Second, we establish lower bounds for the approximation by ReLU neural networks for classes of Sobolev-regular functions. Our results extend recent advances in the approximation theory of ReLU networks to the regime that is most relevant for applications in the numerical analysis of partial differential equations. |
| Tasks | |
| Published | 2019-02-21 |
| URL | http://arxiv.org/abs/1902.07896v1 |
| http://arxiv.org/pdf/1902.07896v1.pdf | |
| PWC | https://paperswithcode.com/paper/error-bounds-for-approximations-with-deep-1 |
| Repo | |
| Framework | |
Universal Bayes consistency in metric spaces
| Title | Universal Bayes consistency in metric spaces |
| Authors | Steve Hanneke, Aryeh Kontorovich, Sivan Sabato, Roi Weiss |
| Abstract | We show that a recently proposed 1-nearest-neighbor-based multiclass learning algorithm is universally strongly Bayes consistent in all metric spaces where such Bayes consistency is possible, making it an optimistically universal Bayes-consistent learner. This is the first learning algorithm known to enjoy this property; by comparison, $k$-NN and its variants are not generally universally Bayes consistent, except under additional structural assumptions, such as an inner product, a norm, finite doubling dimension, or a Besicovitch-type property. The metric spaces in which universal Bayes consistency is possible are the essentially separable ones — a new notion that we define, which is more general than standard separability. The existence of metric spaces that are not essentially separable is independent of the ZFC axioms of set theory. We prove that essential separability exactly characterizes the existence of a universal Bayes-consistent learner for the given metric space. In particular, this yields the first impossibility result for universal Bayes consistency. Taken together, these positive and negative results resolve the open problems posed in Kontorovich, Sabato, Weiss (2017). |
| Tasks | |
| Published | 2019-06-24 |
| URL | https://arxiv.org/abs/1906.09855v2 |
| https://arxiv.org/pdf/1906.09855v2.pdf | |
| PWC | https://paperswithcode.com/paper/universal-bayes-consistency-in-metric-spaces |
| Repo | |
| Framework | |
A unified view of likelihood ratio and reparameterization gradients and an optimal importance sampling scheme
| Title | A unified view of likelihood ratio and reparameterization gradients and an optimal importance sampling scheme |
| Authors | Paavo Parmas, Masashi Sugiyama |
| Abstract | Reparameterization (RP) and likelihood ratio (LR) gradient estimators are used throughout machine and reinforcement learning; however, they are usually explained as simple mathematical tricks without providing any insight into their nature. We use a first principles approach to explain LR and RP, and show a connection between the two via the divergence theorem. The theory motivated us to derive optimal importance sampling schemes to reduce LR gradient variance. Our newly derived distributions have analytic probability densities and can be directly sampled from. The improvement for Gaussian target distributions was modest, but for other distributions such as a Beta distribution, our method could lead to arbitrarily large improvements, and was crucial to obtain competitive performance in evolution strategies experiments. |
| Tasks | |
| Published | 2019-10-14 |
| URL | https://arxiv.org/abs/1910.06419v1 |
| https://arxiv.org/pdf/1910.06419v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-unified-view-of-likelihood-ratio-and |
| Repo | |
| Framework | |
LAGC: Lazily Aggregated Gradient Coding for Straggler-Tolerant and Communication-Efficient Distributed Learning
| Title | LAGC: Lazily Aggregated Gradient Coding for Straggler-Tolerant and Communication-Efficient Distributed Learning |
| Authors | Jingjing Zhang, Osvaldo Simeone |
| Abstract | Gradient-based distributed learning in Parameter Server (PS) computing architectures is subject to random delays due to straggling worker nodes, as well as to possible communication bottlenecks between PS and workers. Solutions have been recently proposed to separately address these impairments based on the ideas of gradient coding, worker grouping, and adaptive worker selection. This paper provides a unified analysis of these techniques in terms of wall-clock time, communication, and computation complexity measures. Furthermore, in order to combine the benefits of gradient coding and grouping in terms of robustness to stragglers with the communication and computation load gains of adaptive selection, novel strategies, named Lazily Aggregated Gradient Coding (LAGC) and Grouped-LAG (G-LAG), are introduced. Analysis and results show that G-LAG provides the best wall-clock time and communication performance, while maintaining a low computational cost, for two representative distributions of the computing times of the worker nodes. |
| Tasks | |
| Published | 2019-05-22 |
| URL | https://arxiv.org/abs/1905.09148v1 |
| https://arxiv.org/pdf/1905.09148v1.pdf | |
| PWC | https://paperswithcode.com/paper/lagc-lazily-aggregated-gradient-coding-for |
| Repo | |
| Framework | |
Optimizing regularized Cholesky score for order-based learning of Bayesian networks
| Title | Optimizing regularized Cholesky score for order-based learning of Bayesian networks |
| Authors | Qiaoling Ye, Arash A. Amini, Qing Zhou |
| Abstract | Bayesian networks are a class of popular graphical models that encode causal and conditional independence relations among variables by directed acyclic graphs (DAGs). We propose a novel structure learning method, annealing on regularized Cholesky score (ARCS), to search over topological sorts, or permutations of nodes, for a high-scoring Bayesian network. Our scoring function is derived from regularizing Gaussian DAG likelihood, and its optimization gives an alternative formulation of the sparse Cholesky factorization problem from a statistical viewpoint, which is of independent interest. We combine global simulated annealing over permutations with a fast proximal gradient algorithm, operating on triangular matrices of edge coefficients, to compute the score of any permutation. Combined, the two approaches allow us to quickly and effectively search over the space of DAGs without the need to verify the acyclicity constraint or to enumerate possible parent sets given a candidate topological sort. The annealing aspect of the optimization is able to consistently improve the accuracy of DAGs learned by local search algorithms. In addition, we develop several techniques to facilitate the structure learning, including pre-annealing data-driven tuning parameter selection and post-annealing constraint-based structure refinement. Through extensive numerical comparisons, we show that ARCS achieves substantial improvements over existing methods, demonstrating its great potential to learn Bayesian networks from both observational and experimental data. |
| Tasks | |
| Published | 2019-04-28 |
| URL | http://arxiv.org/abs/1904.12360v1 |
| http://arxiv.org/pdf/1904.12360v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimizing-regularized-cholesky-score-for |
| Repo | |
| Framework | |
