Sliding-window object detectors that generate bounding-box object predictionsover a dense, regular grid have advanced rapidly and proven popular. Incontrast, modern instance segmentation... | Continue reading
Deep Reinforcement Learning (DRL) has recently spread into a range of domainswithin physics and engineering, with multiple remarkable achievements. Still,much remains to be explored before the... | Continue reading
Artificial Intelligence (AI) has received tremendous attention from academia,industry, and the general public in recent years. The integration of geographyand AI, or GeoAI, provides novel... | Continue reading
European Law now requires AI to be explainable in the context of adversedecisions affecting European Union (EU) citizens. At the same time, it isexpected that there will be increasing instances... | Continue reading
In this note, I quantitatively examine various trends in the lengths ofpublished papers in ACM CSCW from 2000-2018, focusing on several majortransitions in editorial and reviewing policy. The... | Continue reading
Incentive mechanisms play a pivotal role in collecting correct and reliableinformation from self-interested agents. Peer-prediction mechanisms aregame-theoretic mechanisms that incentivize... | Continue reading
Fake news is risky since it has been created to manipulate the readers'opinions and beliefs. In this work, we compared the language of false news tothe real one of real news from an emotional... | Continue reading
Perhaps the biggest hurdle to mankind's expansion throughout the Solar Systemis the prohibitive cost of escaping Earth's gravitational pull. In its manyforms, the space-elevator provides a way... | Continue reading
Social media platforms like Facebook and Twitter permit experiments to beperformed at minimal cost on populations of a size that scientists mightpreviously have dreamt about. For instance, one... | Continue reading
Experimental effort and animal welfare are concerns when exploring theeffects a compound has on an organism. Appropriate methods for extrapolatingchemical effects can further mitigate these... | Continue reading
In general, video games not only prevail in entertainment but also havebecome an alternative methodology for knowledge learning, skill acquisition andassistance for medical treatment as well as... | Continue reading
According to screenwriting theory, turning points (e.g., change of plans,major setback, climax) are crucial narrative moments within a screenplay: theydefine the plot structure, determine its... | Continue reading
Background. Pull requests are a common practice for contributing andreviewing contributions, and are employed both in open-source and industrialcontexts. One of the main goals of code reviews is... | Continue reading
In this paper we consider self-supervised representation learning to improvesample efficiency in reinforcement learning (RL). We propose a forwardprediction objective for simultaneously learning... | Continue reading
To extract the structured representations of open-domain events, Bayesiangraphical models have made some progress. However, these approaches typicallyassume that all words in a document are... | Continue reading
Large language models have a range of beneficial uses: they can assist inprose, poetry, and programming; analyze dataset biases; and more. However,their flexibility and generative capabilities... | Continue reading
We apply BERT to coreference resolution, achieving strong improvements on theOntoNotes (+3.9 F1) and GAP (+11.5 F1) benchmarks. A qualitative analysis ofmodel predictions indicates that,... | Continue reading
Imitation learning is an effective alternative approach to learn a policywhen the reward function is sparse. In this paper, we consider a challengingsetting where an agent and an expert use... | Continue reading
Text summarization aims at compressing long documents into a shorter formthat conveys the most important parts of the original document. Despiteincreased interest in the community and notable... | Continue reading
We study the problem of drawing a dynamic graph, where each vertex appears inthe graph at a certain time and remains in the graph for a fixed amount oftime, called the window size. This defines... | Continue reading
The suffix array is a fundamental data structure for many applications thatinvolve string searching and data compression. Designing time/space-efficientsuffix array construction algorithms has... | Continue reading
Motivated by the recent progresses on machine learning-based models thatlearn artistic styles, in this paper we focus on the problem of poemgeneration. This is a challenging task in which the... | Continue reading
A conversational agent (chatbot) is a piece of software that is able tocommunicate with humans using natural language. Modeling conversation is animportant task in natural language processing... | Continue reading
CRF has been used as a powerful model for statistical sequence labeling. Forneural sequence labeling, however, BiLSTM-CRF does not always lead to betterresults compared with BiLSTM-softmax local... | Continue reading
The question of generalization in machine learning---how algorithms are ableto learn predictors from a training sample to make accurate predictionsout-of-sample---is revisited in light of the... | Continue reading
BERT-based architectures currently give state-of-the-art performance on manyNLP tasks, but little is known about the exact mechanisms that contribute toits success. In the current work, we focus... | Continue reading
Back-translation based approaches have recently lead to significant progressin unsupervised sequence-to-sequence tasks such as machine translation or styletransfer. In this work, we extend the... | Continue reading
A common representation of information about relations of objects andattributes in knowledge domains are data-tables. The structure of suchinformation can be analysed using Formal Concept... | Continue reading
Current 3D object detection methods are heavily influenced by 2D detectors.In order to leverage architectures in 2D detectors, they often convert 3D pointclouds to regular grids (i.e., to voxel... | Continue reading
Face Super-Resolution (SR) is a subfield of the SR domain that specificallytargets the reconstruction of face images. The main challenge of face SR is torestore essential facial features without... | Continue reading
We present a 3D Convolutional Neural Networks (CNNs) based single shotdetector for spatial-temporal action detection tasks. Our model includes: (1)two short-term appearance and motion streams,... | Continue reading
Registration of 3D point clouds is a fundamental task in several applicationsof robotics and computer vision. While registration methods such as iterativeclosest point and variants are very... | Continue reading
Access to the system resources. The current access control systems face manyproblems, such as the presence of the third-party, inefficiency, and lack ofprivacy. These problems can be addressed... | Continue reading
We present a new construction for the Hodge operator for differentialmanifolds based on a Fourier (Berezin)-integral representation. We find asimple formula for the Hodge dual of the wedge... | Continue reading
We focus on the task of unsupervised lemmatization, i.e. grouping togetherinflected forms of one word under one label (a lemma) without the use ofannotated training data. We propose to perform... | Continue reading
Dialogue quality assessment is crucial for evaluating dialogue agents. Anessential factor of high-quality dialogues is coherence - what makes dialogueutterances a whole. This paper proposes a... | Continue reading
Mathematicians manipulate sets with confidence almost every day, rarelymaking mistakes. Few of us, however, could accurately quote what are oftenreferred to as "the" axioms of set theory. This... | Continue reading
Electronic health records (EHRs) stored in hospital information systemscompletely reflect the patients' diagnosis and treatment processes, which areessential to clinical data mining. Chinese... | Continue reading
Bidirectional typing combines two modes of typing: type checking, whichchecks that a program satisfies a known type, and type synthesis, whichdetermines a type from the program. Using checking... | Continue reading
Competitive debaters often find themselves facing a challenging task -- howto debate a topic they know very little about, with only minutes to prepare,and without access to books or the... | Continue reading
Non-profits and the media claim there is a radicalization pipeline onYouTube. Its content creators would sponsor fringe ideas, and its recommendersystem would steer users towards edgier content.... | Continue reading
Entity alignment is the task of linking entities with the same real-worldidentity from different knowledge graphs (KGs), which has been recentlydominated by embedding-based methods. Such... | Continue reading
With increasing information from social media, there are more and more videosavailable. Therefore, the ability to reason on a video is important anddeserves to be discussed. TheDialog System... | Continue reading
In this work, we present X-SQL, a new network architecture for the problem ofparsing natural language to SQL query. X-SQL proposes to enhance the structuralschema representation with the... | Continue reading
The fluid flow and heat transfer problems encountered in industryapplications span into different scales and there are different numericalmethods for different scales problems. It is not... | Continue reading
For neural sequence model training, maximum likelihood (ML) has been commonlyadopted to optimize model parameters with respect to the correspondingobjective. However, in the case of sequence... | Continue reading
In this paper, we study the problem of short sentence ranking for questionanswering. In order to get best score for all the sentences when given a query.We compute the representation for all the... | Continue reading
The Shapley value has become a popular method to attribute the prediction ofa machine-learning model on an input to its base features. The Shapley value[1] is known to be the unique method that... | Continue reading