Distribution Service (DDS) is a realtime peer-to-peer protocol that serves asa scalable middleware between distributed networked systems found in manyIndustrial IoT domains such as automotive,... | Continue reading
Self-supervision techniques have allowed neural language models to advancethe frontier in Natural Language Understanding. However, existingself-supervision techniques operate at the word-form... | Continue reading
Semantically non-compositional phrases constitute an intriguing researchtopic in Natural Language Processing. Semantic non-compositionality --thesituation when the meaning of a phrase cannot be... | Continue reading
For languages with simple morphology, such as English, automatic annotationpipelines such as spaCy or Stanford's CoreNLP successfully serve projects inacademia and the industry. For many... | Continue reading
In this paper, we proposed a deep learning-based end-to-end method on thedomain specified automatic term extraction (ATE), it considers possible termspans within a fixed length in the sentence... | Continue reading
We present XCMRC, the first public cross-lingual language understanding (XLU)benchmark which aims to test machines on their cross-lingual readingcomprehension ability. To be specific, XCMRC is a... | Continue reading
In this paper, we propose a novel end-to-end framework called KBRD, whichstands for Knowledge-Based Recommender Dialog System. It integrates therecommender system and the dialog generation... | Continue reading
Most existing approaches to disfluency detection heavily rely onhuman-annotated data, which is expensive to obtain in practice. To tackle thetraining data bottleneck, we investigate methods for... | Continue reading
Taking word sequences as the input, typical named entity recognition (NER)models neglect errors from pre-processing (e.g., tokenization). However, theseerrors can influence the model performance... | Continue reading
This paper focuses on "tracing player knowledge" in educational games.Specifically, given a set of concepts or skills required to master a game, thegoal is to estimate the likelihood with which... | Continue reading
Melody generation from lyrics has been a challenging research issue in thefield of artificial intelligence and music, which enables to learn and discoverlatent relationship between interesting... | Continue reading
Topological superconductivity hosts exotic quasi-particle excitationsincluding Majorana bound states which hold promise for fault-tolerant quantumcomputing. The theory predicts emergence of... | Continue reading
This paper presents Knowledge-Based Reinforcement Learning (KB-RL) as amethod that combines a knowledge-based approach and a reinforcement learning(RL) technique into one method for intelligent... | Continue reading
Understanding and reasoning about physics is an important ability ofintelligent agents. We develop the PHYRE benchmark for physical reasoning thatcontains a set of simple classical mechanics... | Continue reading
In September 2019, voters for the election at the Parliament of the city ofMoscow will be allowed to use an internet voting system. The source code of ithas been made available for public... | Continue reading
Natural language understanding (NLU) of text is a fundamental challenge inAI, and it has received significant attention throughout the history of NLPresearch. This primary goal has been studied... | Continue reading
Knowledge is a network of interconnected concepts. Yet, precisely how thetopological structure of knowledge constrains its acquisition remains unknown,hampering the development of learning... | Continue reading
Learning embeddings of entities and relations existing in knowledge basesallows the discovery of hidden patterns not captured in the original data. Inthis paper, we propose a novel embedding... | Continue reading
Deep reinforcement learning has learned to play many games well, but failedon others. To better characterize the modes and reasons of failure of deepreinforcement learners, we test the widely... | Continue reading
We focus on multi-turn response selection in a retrieval-based dialog system.In this paper, we utilize the powerful pre-trained language modelBi-directional Encoder Representations from... | Continue reading
The discovery of photonic crystals thirty years ago, in conjunction withresearch advances in plasmonics and metamaterials, has inspired the concept ofdecameter scale metasurfaces, coined seismic... | Continue reading
Explainable AI is an important area of research within which ExplainablePlanning is an emerging topic. In this paper, we argue that ExplainablePlanning can be designed as a service -- that is,... | Continue reading
Embodied Question Answering (EQA) is a recently proposed task, where an agentis placed in a rich 3D environment and must act based solely on its egocentricinput to answer a given question. The... | Continue reading
Open source projects have an increasing importance on modern softwaredevelopment. For this reason, these projects, as usual with commercial softwareprojects, should make use of promotion... | Continue reading
Every day, thousands of customers post questions on Amazon product pages.After some time, if they are fortunate, a knowledgeable customer might answertheir question. Observing that many... | Continue reading
We introduce Differentiable Reasoning (DR), a novel semi-supervised learningtechnique which uses relational background knowledge to benefit from unlabeleddata. We apply it to the Semantic Image... | Continue reading
Holm 15A, the brightest cluster galaxy (BCG) of the cool-core cluster Abell85, has an ultra-diffuse central region, 2 mag fainter than the faintestdepleted core of any early-type galaxy (ETG)... | Continue reading
Consistent and reproducible evaluation of Deep Reinforcement Learning (DRL)is not straightforward. In the Arcade Learning Environment (ALE), small changesin environment parameters such as... | Continue reading
Game development is a complex task involving multiple disciplines andtechnologies. Developers and researchers alike have suggested that AI-drivengame design assistants may improve developer... | Continue reading
Let $ψ:\mathbb{N}\to\mathbb{R}_{\ge0}$ be an arbitrary function from thepositive integers to the non-negative reals. Consider the set $\mathcal{A}$ ofreal numbers $α$ for which there are... | Continue reading
In this research, we provide a comprehensive empirical summary of the PythonPackage Repository, PyPI, including both package metadata and source codecovering 178,592 packages, 1,745,744... | Continue reading
We critically appraise the recent interest in out-of-distribution (OOD)detection, questioning the practical relevance of existing benchmarks. Whilethe currently prevalent trend is to consider... | Continue reading
Given a video of a person in action, we can easily guess the 3D future motionof the person. In this work, we present perhaps the first approach forpredicting a future 3D mesh model sequence of a... | Continue reading
Ethereum has emerged as a dynamic platform for exchanging cryptocurrencytokens. While token crowdsales cannot simultaneously guarantee buyers bothcertainty of valuation and certainty of... | Continue reading
Small low-mass stars are favourable targets for the detection of rockyhabitable planets. In particular, planetary systems in the solar neighbourhoodare interesting and suitable for precise... | Continue reading
We explain how to design RISC-V shellcodes capable of running arbitrary code,whose ASCII binary representation use only letters a-zA-Z, digits 0-9, andeither of the three characters: #, /, '. | Continue reading
Cloud service providers offer a low-cost and convenient solution to hostunstructured data. However, cloud services act as third-party solutions and donot provide control of the data to users.... | Continue reading
Voice-enabled interactions provide more human-like experiences in manypopular IoT systems. Cloud-based speech analysis services extract usefulinformation from voice input using speech... | Continue reading
We present a corpus of Finnish news articles with a manually prepared namedentity annotation. The corpus consists of 953 articles (193,742 word tokens)with six named entity classes... | Continue reading
Explainability of deep learning systems is a vital requirement for manyapplications. However, it is still an unsolved problem. Recent self-attentionbased models for natural language processing,... | Continue reading
Despite recent progress in computer vision, finegrained interpretation ofsatellite images remains challenging because of a lack of labeled trainingdata. To overcome this limitation, we construct... | Continue reading
Several companies are re-architecting their monolithic information systemswith microservices. However, many companies migrated without experience onmicroservices, mainly learning how to migrate... | Continue reading
We use optimism to introduce generic asymptotically optimal reinforcementlearning agents. They achieve, with an arbitrary finite or compact class ofenvironments, asymptotically optimal behavior.... | Continue reading
We present an analysis into the inner workings of Convolutional NeuralNetworks (CNNs) for processing text. CNNs used for computer vision can beinterpreted by projecting filters into image space,... | Continue reading
Security analysts need to classify, search and correlate numerous images.Automatic classification tools improve the efficiency of such tasks. However,no open-source and turnkey library was found... | Continue reading
Advances in imaging technologies, combined with inexpensive storage, have ledto an explosion in the volume of publicly available neuroimaging datasets.Effective analyses of these images hold the... | Continue reading
We introduce a language-agnostic evolutionary technique for automaticallyextracting chunks from dependency treebanks. We evaluate these chunks on anumber of morphosyntactic tasks, namely POS... | Continue reading
There has been huge progress in speech recognition over the last severalyears. Tasks once thought extremely difficult, such as SWITCHBOARD, nowapproach levels of human performance. The MALACH... | Continue reading