Papers.

Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees.

Abstract 16231: Machine Learning of Remodeled Ventricular Action Potentials and Long-Term Follow-Up of Ventricular Arrhythmias

Abstract 14675: Developing Convolutional Neural Networks for Deep Learning of Ventricular Action Potentials to Predict Risk for Ventricular Arrhythmias

A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning.

A Meta-Analysis of Overfitting in Machine Learning

A brief course in modern math for programmers

Programmers with the classical software engineering background need to learn more mathematics these days. The world is changing; we have to change, too. This book is dedicated to covering the issues that until recently were not very popular in software engineering: logic, monoids, algebraic structures, categories, and monads. All these topics are explained in the book, with no assumptions about the reader's educational background, with many examples. Most of the examples and explanations use two popular programming languages, JavaScript and Scala. The book has no theorems and almost no proofs. The purpose of the book is to expand the reader's imagination and to open the gates to the beautiful world of mathematics—at the same time keeping in mind the practical usability of its ideas and notions in our daily coding practice.

Do You Have a Conversational Interface?

Essay arguing that conversational interfaces - chatbots and voice assistants - are becoming a primary channel between companies and customers, and setting out what managers must decide about platform choice, natural-language capability, and organizational readiness before building one.

Towards Progress in Theories of Language Sound Structure

From ‘Solved Problems’ to New Challenges: A Report on LDC Activities

Introducing NIEUW: Novel Incentives and Workflows for Eliciting Linguistic Data

First Dihard Challenge Evaluation Plan

Ptolemy's Britain and Ireland: A New Digital Reconstruction

Abstract. In this paper, we expand application of our mathematical methods for translating ancient coordinates from the classical Geography by Claudius Ptolemy into modern coordinates from India and Arabia to Britain and Ireland, historically important islands on the periphery of the ancient Roman Empire. The methods include triangulation and flocking with subsequent Bayesian correction. The results of our work can be conveniently visualized in modern GIS tools, such as ArcGIS, QGIS, and Google Earth. The enhancements we have made include a novel technique for handling tentatively identified points. We compare the precision of reconstruction achieved for Ptolemy's Britain and Ireland with the precisions that we had computed earlier for his India before the Ganges and three provinces of Arabia. We also provide improved validation and comparison amongst the methods applied. We compare our results with the prior work, while utilizing knowledge from such important ancient sources as the Antonine Itinerary, Tabula Peutingeriana, and the Ravenna Cosmography. The new digital reconstruction of Claudius Ptolemy's Britain and Ireland presented in this paper, along with the accompanying linguistic analysis of ancient toponyms, contributes to improvement of understanding of our cultural cartographic heritage by making it easier to study the ancient world using the popular and accessible GIS programs.

Resilience, buoyancy and grit

The promise of seamless mobility

Integrating Voice Quality Cues in the Pitch Perception of Speech and Non-speech Utterances

Real-time Money Routing by Trusting Strangers with your Funds

A Blockchain Consensus Protocol With Horizontal Scalability

From Hyperlinks to Hyperties

Information structure and prosodic prominence: how does sentence final particle affect Cantonese intonation?

Deep Neural Networks for Emotion Recognition Combining Audio and Transcripts

In this paper, we propose to improve emotion recognition by combining acoustic information and conversation transcripts. On the one hand, an LSTM network was used to detect emotion from acoustic features like f0, shimmer, jitter, MFCC, etc. On the other hand, a multi-resolution CNN was used to detect emotion from word sequences. This CNN consists of several parallel convolutions with different kernel sizes to exploit contextual information at different levels. A temporal pooling layer aggregates the hidden representations of different words into a unique sequence level embedding, from which we computed the emotion posteriors. We optimized a weighted sum of classification and verification losses. The verification loss tries to bring embeddings from the same emotions closer while separating embeddings from different emotions. We also compared our CNN with state-of-the-art text-based hand-crafted features (e-vector). We evaluated our approach on the USC-IEMOCAP dataset as well as the dataset consisting of US English telephone speech. In the former, we used human-annotated transcripts while in the latter, we used ASR transcripts. The results showed fusing audio and transcript information improved unweighted accuracy by relative 24% for IEMOCAP and relative 3.4% for the telephone data compared to a single acoustic system.

GlobalTIMIT: Acoustic-Phonetic Datasets for the World’s Languages

Punctuation Prediction Model for Conversational Speech

An ASR system usually does not predict any punctuation or capitalization. Lack of punctuation causes problems in result presentation and confuses both the human reader andoff-the-shelf natural language processing algorithms. To overcome these limitations, we train two variants of Deep Neural Network (DNN) sequence labelling models - a Bidirectional Long Short-Term Memory (BLSTM) and a Convolutional Neural Network (CNN), to predict the punctuation. The models are trained on the Fisher corpus which includes punctuation annotation. In our experiments, we combine time-aligned and punctuated Fisher corpus transcripts using a sequence alignment algorithm. The neural networks are trained on Common Web Crawl GloVe embedding of the words in Fisher transcripts aligned with conversation side indicators and word time infomation. The CNNs yield a better precision and BLSTMs tend to have better recall. While BLSTMs make fewer mistakes overall, the punctuation predicted by the CNN is more accurate - especially in the case of question marks. Our results constitute significant evidence that the distribution of words in time, as well as pre-trained embeddings, can be useful in the punctuation prediction task.

Foafing the Music: Bridging the Semantic Gap in Music Recommendation

ZemPod: A Semantic Web Approach to Podcasting

A Practical q -Gram Index for Text Retrieval Allowing Errors

E2E NLG Challenge Submission: Towards Controllable Generation of Diverse Natural Language

Continuous Learning in a Hierarchical Multiscale Neural Network

Semantically Equivalent Adversarial Rules for Debugging NLP models

Complex machine learning models for NLP are often brittle, making different predictions for input instances that are extremely similar semantically. To automatically detect this behavior for individual instances, we present semantically equivalent adversaries (SEAs) – semantic-preserving perturbations that induce changes in the model’s predictions. We generalize these adversaries into semantically equivalent adversarial rules (SEARs) – simple, universal replacement rules that induce adversaries on many instances. We demonstrate the usefulness and flexibility of SEAs and SEARs by detecting bugs in black-box state-of-the-art models for three domains: machine comprehension, visual question-answering, and sentiment analysis. Via user studies, we demonstrate that we generate high-quality local adversaries for more instances than humans, and that SEARs induce four times as many mistakes as the bugs discovered by human experts. SEARs are also actionable: retraining models using data augmentation significantly reduces bugs, while maintaining accuracy.

Universal Language Model Fine-tuning for Text Classification

Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch. We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce techniques that are key for fine-tuning a language model. Our method significantly outperforms the state-of-the-art on six text classification tasks, reducing the error by 18-24% on the majority of datasets. Furthermore, with only 100 labeled examples, it matches the performance of training from scratch on 100 times more data. We open-source our pretrained models and code.

A La Carte Embedding: Cheap but Effective Induction of Semantic Feature Vectors.

Cross-lingual Transfer Learning for Multilingual Task Oriented Dialog

One of the first steps in the utterance interpretation pipeline of many task-oriented conversational AI systems is to identify user intents and the corresponding slots. Since data collection for machine learning models for this task is time-consuming, it is desirable to make use of existing data in a high-resource language to train models in low-resource languages. However, development of such models has largely been hindered by the lack of multilingual training data. In this paper, we present a new data set of 57k annotated utterances in English (43k), Spanish (8.6k) and Thai (5k) across the domains weather, alarm, and reminder. We use this data set to evaluate three different cross-lingual transfer methods: (1) translating the training data, (2) using cross-lingual pre-trained embeddings, and (3) a novel method of using a multilingual machine translation encoder as contextual word representations. We find that given several hundred training examples in the the target language, the latter two methods outperform translating the training data. Further, in very low-resource settings, multilingual contextual word representations give better results than using cross-lingual static embeddings. We also compare the cross-lingual methods to using monolingual resources in the form of contextual ELMo representations and find that given just small amounts of target language data, this method outperforms all cross-lingual methods, which highlights the need for more sophisticated cross-lingual methods.

360° Stance Detection

SystemT: Declarative Text Understanding for Enterprise

The rise of enterprise applications over unstructured and semi-structured documents poses new challenges to text understanding systems across multiple dimensions. We present SystemT, a declarative text understanding system that addresses these challenges and has been deployed in a wide range of enterprise applications. We highlight the design considerations and decisions behind SystemT in addressing the needs of the enterprise setting. We also summarize the impact of SystemT on business and education.

Embedding Multimodal Relational Data for Knowledge Base Completion

Representing entities and relations in an embedding space is a well-studied approach for machine learning on relational data. Existing approaches, however, primarily focus on simple link structure between a finite set of entities, ignoring the variety of data types that are often used in knowledge bases, such as text, images, and numerical values. In this paper, we propose multimodal knowledge base embeddings (MKBE) that use different neural encoders for this variety of observed data, and combine them with existing relational models to learn embeddings of the entities and multimodal data. Further, using these learned embedings and different neural decoders, we introduce a novel multimodal imputation model to generate missing multimodal values, like text and images, from information in the knowledge base. We enrich existing relational datasets to create two novel benchmarks that contain additional information such as textual descriptions and images of the original entities. We demonstrate that our models utilize this additional information effectively to provide more accurate link prediction, achieving state-of-the-art results with a considerable gap of 5-7% over existing methods. Further, we evaluate the quality of our generated multimodal values via a user study.

Semantic Parsing for Task Oriented Dialog using Hierarchical Representations

Task oriented dialog systems typically first parse user utterances to semantic frames comprised of intents and slots. Previous work on task oriented intent and slot-filling work has been restricted to one intent per query and one slot label per token, and thus cannot model complex compositional requests. Alternative semantic parsing systems have represented queries as logical forms, but these are challenging to annotate and parse. We propose a hierarchical annotation scheme for semantic parsing that allows the representation of compositional queries, and can be efficiently and accurately parsed by standard constituency parsing models. We release a dataset of 44k annotated queries (http://fb.me/semanticparsingdialog), and show that parsing models outperform sequence-to-sequence approaches on this dataset.

Interpretation of Natural Language Rules in Conversational Machine Reading

Most work in machine reading focuses on question answering problems where the answer is directly expressed in the text to read. However, many real-world question answering problems require the reading of text not because it contains the literal answer, but because it contains a recipe to derive an answer together with the reader’s background knowledge. One example is the task of interpreting regulations to answer “Can I...?” or “Do I have to...?” questions such as “I am working in Canada. Do I have to carry on paying UK National Insurance?” after reading a UK government website about this topic. This task requires both the interpretation of rules and the application of background knowledge. It is further complicated due to the fact that, in practice, most questions are underspecified, and a human assistant will regularly have to ask clarification questions such as “How long have you been working abroad?” when the answer cannot be directly derived from the question and text. In this paper, we formalise this task and develop a crowd-sourcing strategy to collect 37k task instances based on real-world rules and crowd-generated questions and scenarios. We analyse the challenges of this task and assess its difficulty by evaluating the performance of rule-based and machine-learning baselines. We observe promising results when no background knowledge is necessary, and substantial room for improvement whenever background knowledge is needed.

DeepHeart: Semi-Supervised Sequence Learning for Cardiovascular Risk Prediction

Anchors: High-Precision Model-Agnostic Explanations

We introduce a novel model-agnostic system that explains the behavior of complex models with high-precision rules called anchors, representing local, "sufficient" conditions for predictions. We propose an algorithm to efficiently compute these explanations for any black-box model with high-probability guarantees. We demonstrate the flexibility of anchors by explaining a myriad of different models for different domains and tasks. In a user study, we show that anchors enable users to predict how a model would behave on unseen instances with less effort and higher precision, as compared to existing linear explanations or no explanations.

DIFF: a relational interface for large-scale data explanation

A range of explanation engines assist data analysts by performing feature selection over increasingly high-volume and high-dimensional data, grouping and highlighting commonalities among data points. While useful in diverse tasks such as user behavior analytics, operational event processing, and root cause analysis, today's explanation engines are designed as standalone data processing tools that do not interoperate with traditional, SQL-based analytics workflows; this limits the applicability and extensibility of these engines. In response, we propose the DIFF operator, a relational aggregation operator that unifies the core functionality of these engines with declarative relational query processing. We implement both single-node and distributed versions of the DIFF operator in MB SQL, an extension of MacroBase, and demonstrate how DIFF can provide the same semantics as existing explanation engines while capturing a broad set of production use cases in industry, including at Microsoft and Facebook. Additionally, we illustrate how this declarative approach to data explanation enables new logical and physical query optimizations. We evaluate these optimizations on several real-world production applications, and find that DIFF in MB SQL can outperform state-of-the-art engines by up to an order of magnitude.

Locality-sensitive hashing for earthquake detection

Moment-based quantile sketches for efficient high cardinality aggregation queries

Filter before you parse

Axiomatic Foundations and Algorithms for Deciding Semantic Equivalences of SQL Queries

Deciding the equivalence of SQL queries is a fundamental problem in data management. As prior work has mainly focused on studying the theoretical limitations of the problem, very few implementations for checking such equivalences exist. In this paper, we present a new formalism and implementation for reasoning about the equivalences of SQL queries. Our formalism, U-semiring, extends SQL's semiring semantics with unbounded summation and duplicate elimination. U-semiring is defined using only very few axioms and can thus be easily implemented using proof assistants such as Lean for automated query reasoning.

Dhalion in Action: Automatic Management of Streaming Applications

In a world where organizations are being inundated with data from various sources, analyzing data and gaining actionable insights in real-time has become a key service differentiator. A crucial challenge in these environments is the complexity of configuring, managing and deploying long-running streaming applications.

Exploring patient information needs in type 2 diabetes: A cross sectional study of questions

Dysprosody markers in PPA (P3.194)

Validation of a software-based clinical trial matching platform for oncology using comprehensive clinical information.

Linear Algebraic Structure of Word Senses, with Applications to Polysemy.

Corpus Phonetics

Identity by Any Other Name

The many meanings of identity across databases, distributed systems and immutable data.