Detecting and quantifying anomalies in urban traffic is critical for real-time alerting or re-routing in the short run and urban planning in the long run. We describe a two-step framework that achieves these two goals in a robust, fast, online, and unsupervised manner. First, we adapt stable principal component pursuit to detect anomalies for each road segment. This allows us to pinpoint traffic anomalies early and precisely in space. Then we group the road-level anomalies across time and space into meaningful anomaly events using a simple graph expansion procedure. These events can be easily clustered, visualized, and analyzed by urban planners. We demonstrate the effectiveness of our system using 7 weeks of anonymized and aggregated cellular location data in Dallas-Fort Worth. We suggest potential opportunities for urban planners and policy makers to use our methodology to make informed changes. These applications include real-time re-routing of traffic in response to abnormally high traffic, or identifying candidates for high-impact infrastructure projects.

Charts are an excellent way to convey patterns and trends in data, but they do not facilitate further modeling of the data or close inspection of individual data points. We present a fully automated system for extracting the numerical values of data points from images of scatter plots. We use deep learning techniques to identify the key components of the chart, and optical character recognition together with robust regression to map from pixels to the coordinate system of the chart. We focus on scatter plots with linear scales, which already have several interesting challenges. Previous work has done fully automatic extraction for other types of charts, but to our knowledge this is the first approach that is fully automatic for scatter plots. Our method performs well, achieving successful data extraction on 89% of the plots in our test set.

In this paper, we introduce the first and largest Hindi text corpus, named BHAAV, which means emotions in Hindi, for analyzing emotions that a writer expresses through his characters in a story, as perceived by a narrator/reader. The corpus consists of 20,304 sentences collected from 230 different short stories spanning across 18 genres such as Inspirational and Mystery. Each sentence has been annotated into one of the five emotion categories - anger, joy, suspense, sad, and neutral, by three native Hindi speakers with at least ten years of formal education in Hindi. We also discuss challenges in the annotation of low resource languages such as Hindi, and discuss the scope of the proposed corpus along with its possible uses. We also provide a detailed analysis of the dataset and train strong baseline classifiers reporting their performances.

Learning how to predict future events from patterns of past events is difficult when the set of possible event types is large. Training an unrestricted neural model might overfit to spurious patterns. To exploit domain-specific knowledge of how past events might affect an event's present probability, we propose using a temporal deductive database to track structured facts over time. Rules serve to prove facts from other facts and from past events. Each fact has a time-varying state---a vector computed by a neural net whose topology is determined by the fact's provenance, including its experience of past events. The possible event types at any time are given by special facts, whose probabilities are neurally modeled alongside their states. In both synthetic and real-world domains, we show that neural probabilistic models derived from concise Datalog programs improve prediction by encoding appropriate domain knowledge in their architecture.

We investigate the approximation of path functionals. In particular, we advocate the use of the Karhunen-Loève expansion, the continuous analogue of Principal Component Analysis, to extract relevant information from the image of a functional. Having accurate estimate of functionals is of paramount importance in the context of exotic derivatives pricing, as presented in the practical applications. Specifically, we show how a simulation-based procedure, which we call the Karhunen-Loève Monte Carlo (KLMC) algorithm, allows fast and efficient computation of the price of path-dependent options. We also explore the path signature as an alternative tool to project both paths and functionals.

Complaining is a basic speech act regularly used in human and computer mediated communication to express a negative mismatch between reality and expectations in a particular situation. Automatically identifying complaints in social media is of utmost importance for organizations or brands to improve the customer experience or in developing dialogue systems for handling and responding to complaints. In this paper, we introduce the first systematic analysis of complaints in computational linguistics. We collect a new annotated data set of written complaints expressed in English on Twitter.\footnote{Data and code is available here: \url{this https URL}} We present an extensive linguistic analysis of complaining as a speech act in social media and train strong feature-based and neural models of complaints across nine domains achieving a predictive performance of up to 79 F1 using distant supervision.

Hashtags are often employed on social media and beyond to add metadata to a textual utterance with the goal of increasing discoverability, aiding search, or providing additional semantics. However, the semantic content of hashtags is not straightforward to infer as these represent ad-hoc conventions which frequently include multiple words joined together and can include abbreviations and unorthodox spellings. We build a dataset of 12,594 hashtags split into individual segments and propose a set of approaches for hashtag segmentation by framing it as a pairwise ranking problem between candidate segmentations. Our novel neural approaches demonstrate 24.6% error reduction in hashtag segmentation accuracy compared to the current state-of-the-art method. Finally, we demonstrate that a deeper understanding of hashtag semantics obtained through segmentation is useful for downstream applications such as sentiment analysis, for which we achieved a 2.6% increase in average recall on the SemEval 2017 sentiment analysis dataset.

Photographs, taken by field scientists, tourists, automated cameras, and incidental photographers, are the most abundant source of data on wildlife today. Wildbook is an autonomous computational system that starts from massive collections of images and, by detecting various species of animals and identifying individuals, combined with sophisticated data management, turns them into high resolution information database, enabling scientific inquiry, conservation, and citizen science. We have built Wildbooks for whales (flukebook.org), sharks (whaleshark.org), two species of zebras (Grevy's and plains), and several others. In January 2016, Wildbook enabled the first ever full species (the endangered Grevy's zebra) census using photographs taken by ordinary citizens in Kenya. The resulting numbers are now the official species census used by IUCN Red List: this http URL In 2016, Wildbook partnered up with WWF to build Wildbook for Sea Turtles, Internet of Turtles (IoT), as well as systems for seals and lynx. Most recently, we have demonstrated that we can now use publicly available social media images to count and track wild animals. In this paper we present and discuss both the impact and challenges that the use of crowdsourced images can have on wildlife conservation.

Previous work has found strong links between the choice of social media images and users' emotions, demographics and personality traits. In this study, we examine which attributes of profile and posted images are associated with depression and anxiety of Twitter users. We used a sample of 28,749 Facebook users to build a language prediction model of survey-reported depression and anxiety, and validated it on Twitter on a sample of 887 users who had taken anxiety and depression surveys. We then applied it to a different set of 4,132 Twitter users to impute language-based depression and anxiety labels, and extracted interpretable features of posted and profile pictures to uncover the associations with users' depression and anxiety, controlling for demographics. For depression, we find that profile pictures suppress positive emotions rather than display more negative emotions, likely because of social media self-presentation biases. They also tend to show the single face of the user (rather than show her in groups of friends), marking increased focus on the self, emblematic for depression. Posted images are dominated by grayscale and low aesthetic cohesion across a variety of image features. Profile images of anxious users are similarly marked by grayscale and low aesthetic cohesion, but less so than those of depressed users. Finally, we show that image features can be used to predict depression and anxiety, and that multitask learning that includes a joint modeling of demographics improves prediction performance. Overall, we find that the image attributes that mark depression and anxiety offer a rich lens into these conditions largely congruent with the psychological literature, and that images on Twitter allow inferences about the mental health status of users.

Visual speech recognition (VSR) is the task of recognizing spoken language from video input only, without any audio. VSR has many applications as an assistive technology, especially if it could be deployed in mobile devices and embedded systems. The need of intensive computational resources and large memory footprint are two of the major obstacles in developing neural network models for VSR in a resource constrained environment. We propose a novel end-to-end deep neural network architecture for word level VSR called MobiVSR with a design parameter that aids in balancing the model's accuracy and parameter count. We use depthwise-separable 3D convolution for the first time in the domain of VSR and show how it makes our model efficient. MobiVSR achieves an accuracy of 73\% on a challenging Lip Reading in the Wild dataset with 6 times fewer parameters and 20 times lesser memory footprint than the current state of the art. MobiVSR can also be compressed to 6 MB by applying post training quantization.

The Third Generation Partnership Project (3GPP) released its first 5G security specifications in March 2018. This paper reviews the 5G security architecture, requirements and main processes and evaluates them in the context of known and new protocol exploits. Although the security has been enhanced when compared to previous generations to tackle known protocol exploits, our analysis identifies some potentially unrealistic system assumptions that are critical for security as well as a number protocol edge cases that could render 5G systems vulnerable to adversarial attacks. For example, null encryption and null authentication are supported and can be used in valid system configurations, and certain key security functions are still left outside of the scope of the specifications. Moreover, the prevention of pre-authentcation message exploits appears to rely on the implicit assumption of impractical carrier and roaming agreements and the management of public keys from all global operators. In parallel, existing threats such as International Mobile Subscriber Identity (IMSI) catchers are prevented only if the serving network enforces optional security features and if the UE knows the public key of the home network operator. The comparison with 4G LTE protocol exploits reveals that the 5G security specifications, as of Release 15, do not fully address the user privacy and network availability concerns, where one edge case can compromise the privacy, security and availability of 5G users and services.

We reduce the task of (span-based) PropBank-style semantic role labeling (SRL) to syntactic dependency parsing. Our approach is motivated by our empirical analysis that shows three common syntactic patterns account for over 98% of the SRL annotations for both English and Chinese data. Based on this observation, we present a conversion scheme that packs SRL annotations into dependency tree representations through joint labels that permit highly accurate recovery back to the original format. This representation allows us to train statistical dependency parsers to tackle SRL and achieve competitive performance with the current state of the art. Our findings show the promise of syntactic dependency trees in encoding semantic role relations within their syntactic domain of locality, and point to potential further integration of syntactic methods into semantic role labeling in the future.

We derive sharp bounds for the prices of VIX futures using the full information of S&P 500 smiles. To that end, we formulate the model-free sub/superreplication of the VIX by trading in the S&P 500 and its vanilla options as well as the forward-starting log-contracts. A dual problem of minimizing/maximizing certain risk-neutral expectations is introduced and shown to yield the same value. The classical bounds for VIX futures given the smiles only use a calendar spread of log-contracts on the S&P 500. We analyze for which smiles the classical bounds are sharp and how they can be improved when they are not. In particular, we introduce a family of functionally generated portfolios which often improves the classical bounds while still being tractable; more precisely, determined by a single concave/convex function on the line. Numerical experiments on market data and SABR smiles show that the classical lower bound can be improved dramatically, whereas the upper bound is often close to optimal.

This paper presents our submission to the SemEval 2020 - Task 10 on emphasis selection in written text. We approach this emphasis selection problem as a sequence labeling task where we represent the underlying text with various contextual embedding models. We also employ label distribution learning to account for annotator disagreements. We experiment with the choice of model architectures, trainability of layers, and different contextual embeddings. Our best performing architecture is an ensemble of different models, which achieved an overall matching score of 0.783, placing us 15th out of 31 participating teams. Lastly, we analyze the results in terms of parts of speech tags, sentence lengths, and word ordering.

We provide explicit approximation formulas for VIX futures and options in forward variance models, with particular emphasis on the family of so-called Bergomi models: the one-factor Bergomi model [Bergomi, Smile dynamics II, Risk, 2005], the rough Bergomi model [Bayer, Friz, and Gatheral, Pricing under rough volatility, Quantitative Finance, 16(6):887-904, 2016], and an enhanced version of the rough model that can generate realistic positive skew for VIX smiles -- introduced simultaneously by De Marco [Bachelier World Congress, 2018] and Guyon [Bachelier World Congress, 2018] on the lines of [Bergomi, Smile dynamics III, Risk, 2008], that we refer to as 'mixed rough Bergomi model'. Following the methodology set up in [Gobet and Miri, Weak approximation of averaged diffusion processes. Stochastic Process.\ Appl., 124(1):475-504, 2014], we derive weak approximations for the law of the VIX, leading to option price approximations under the form of explicit combinations of Black-Scholes prices and greeks. As new contributions, we cope with the fractional integration kernel appearing in rough models and treat the case of non-smooth payoffs, so to encompass VIX futures, call and put options. We stress that our approach does not rely on small-time asymptotics nor small-parameter (such as small volatility-of-volatility) asymptotics, and can therefore be applied to any option maturity and a wide range of parameter configurations. Our results are illustrated by several numerical experiments and calibration tests to VIX market data.

With machine learning models being increasingly applied to various decision-making scenarios, people have spent growing efforts to make machine learning models more transparent and explainable. Among various explanation techniques, counterfactual explanations have the advantages of being human-friendly and actionable -- a counterfactual explanation tells the user how to gain the desired prediction with minimal changes to the input. Besides, counterfactual explanations can also serve as efficient probes to the models' decisions. In this work, we exploit the potential of counterfactual explanations to understand and explore the behavior of machine learning models. We design DECE, an interactive visualization system that helps understand and explore a model's decisions on individual instances and data subsets, supporting users ranging from decision-subjects to model developers. DECE supports exploratory analysis of model decisions by combining the strengths of counterfactual explanations at instance- and subgroup-levels. We also introduce a set of interactions that enable users to customize the generation of counterfactual explanations to find more actionable ones that can suit their needs. Through three use cases and an expert interview, we demonstrate the effectiveness of DECE in supporting decision exploration tasks and instance explanations.

Dialogue act (DA) classification has been studied for the past two decades and has several key applications such as workflow automation and conversation analytics. Researchers have used, to address this problem, various traditional machine learning models, and more recently deep neural network models such as hierarchical convolutional neural networks (CNNs) and long short-term memory (LSTM) networks. In this paper, we introduce a new model architecture, directed-acyclic-graph LSTM (DAG-LSTM) for DA classification. A DAG-LSTM exploits the turn-taking structure naturally present in a multi-party conversation, and encodes this relation in its model structure. Using the STAC corpus, we show that the proposed method performs roughly 0.8% better in accuracy and 1.2% better in macro-F1 score when compared to existing methods. The proposed method is generic and not limited to conversation applications.

In December 2017, the Third Generation Partnership Project (3GPP) released the first set of specifications for 5G New Radio (NR), which is currently the most widely accepted 5G cellular standard. 5G NR is expected to replace LTE and previous generations of cellular technology over the next several years, providing higher throughput, lower latency, and a host of new features. Similar to LTE, the 5G NR physical layer consists of several physical channels and signals, most of which are vital to the operation of the network. Unfortunately, like for any wireless technology, disruption through radio jamming is possible. This paper investigates the extent to which 5G NR is vulnerable to jamming and spoofing, by analyzing the physical downlink and uplink control channels and signals. We identify the weakest links in the 5G NR frame, and propose mitigation strategies that should be taken into account during implementation of 5G NR chipsets and base stations.