Search results

We analyze some of the fundamental design challenges that impact the development of a multilingual state-of-the-art named entity transliteration system, including curating bilingual named entity datasets and evaluation of multiple transliteration methods. We empirically evaluate the transliteration task using the traditional weighted finite state transducer (WFST) approach against two neural approaches:

Large scale Natural Language Understanding (NLU) systems are typically trained on large quantities of data, requiring a fast and scalable training strategy. A typical design for NLU systems consists of domain-level NLU modules (domain classifier, intent classifier and named entity recognizer). Hypotheses (NLU interpretations consisting of various intent+slot combinations) from these domain specific modules

Clinical text provides essential information to estimate the acuity of a patient during hospital stays in addition to structured clinical data. In this study, we explore how clinical text can complement a clinical predictive learning task. We leverage an internal medical natural language processing service to perform named entity extraction and negation detection on clinical notes and compose selected entities

Recommendation systems have developed beyond simple matrix factorization to focus on two important sources of information: the temporal order of events (Hidasi et al., 2015) and side (e.g., spatial) information encoded in user and item features (Rendle, 2012). However, state-of-art temporal modeling is often limited by model capacity for long user histories. In addition, meta data are rarely used in generic

Goal-oriented dialogue systems typically rely on components specifically developed for a single task or domain. This limits such systems in two different ways: If there is an update in the task domain, the dialogue system usually needs to be updated or completely re-trained. It is also harder to extend such dialogue systems to different and multiple domains. The dialogue state tracker in conventional dialogue

Fast expansion of natural language functionality of intelligent virtual agents is critical for achieving engaging and informative interactions. However, developing accurate models for new natural language domains is a time and data intensive process. We propose efficient deep neural network architectures that maximally re-use available resources through transfer learning. Our methods are applied for expanding

We consider the problem of labeling points on a fast-moving data stream when only a small number of labeled examples are available. In our setting, incoming points must be processed efficiently and the stream is too large to store in its entirety. We present a semi-supervised learning algorithm for this task. The algorithm maintains a small synopsis of the stream which can be quickly updated as new points

Cloud computing provides on-demand access to IT resources via the Internet. Permissions for these resources are defined by expressive access control policies. This paper presents a formalization of the Amazon Web Services (AWS) policy language and a corresponding analysis tool, called ZELKOVA, for verifying policy properties. ZELKOVA encodes the semantics of policies into SMT, compares behaviors, and verifies

While we have many taxonomies of forecasting methods, the authors present a classification of forecasting problems in modern industrial settings. Such a classification can help decision makers understand what resources to draw upon when facing a particular problem and may lead to more scientific discourse about the relevant data sets for benchmarking forecasting performance.

While the term "deep learning" (DL) has only been coined in the last few years, the techniques it refers to have been in development since the 1950s, namely artificial neural networks (NN or ANN for short). DL has scored major successes in image recognition, natural language processing (e.g. machine translation and speech recognition), and autonomous agents such as Google Deep Mind's AlphaGo. It is often

We present a novel approach to probabilistic time series forecasting that combines state space models with deep learning. By parametrizing a per-time-series linear state space model with a jointly-learned recurrent neural network, our method retains desired properties of state space models such as data efficiency and interpretability, while making use of the ability to learn complex patterns from raw data

We describe SOCKEYE, 1 an open-source sequence-to-sequence toolkit for Neural Machine Translation (NMT). SOCKEYE is a production-ready framework for training and applying models as well as an experimental platform for researchers. Written in Python and built on MXNET, the toolkit offers scalable training and inference for the three most prominent encoder-decoder architectures: attentional recurrent neural

Far-field automatic speech recognition (ASR) is a key enabling technology that allows untethered and natural voice interaction between users and Amazon Echo family of products. A key component in realizing far-field ASR on these products is the suite of audio front-end (AFE) algorithms that helps in mitigating acoustic environmental challenges and thereby improving the ASR performance. In this paper, we

Implementing a Natural Language Processing (NLP) system requires considerable engineering effort: creating data-structures to represent language constructs; reading corpora annotations into these data-structures; applying off-the-shelf NLP tools to augment the text representation; extracting features and training machine learning components; conducting experiments and computing performance statistics; and

Building open domain conversational systems that allow users to have engaging conversations on topics of their choice is a challenging task. Alexa Prize was launched in 2016 to tackle the problem of achieving natural, sustained, coherent and engaging open-domain dialogs. In the second iteration of the competition in 2018, university teams advanced the state of the art by using context in dialog models,

Recent work has made significant progress in improving spatial resolution for pixelwise labeling with Fully Convolutional Network (FCN) framework by employing Dilated/Atrous convolution, utilizing multi-scale features and refining boundaries. In this paper, we explore the impact of global contextual information in semantic segmentation by introducing the Context Encoding Module, which captures the semantic

This repository holds short helper code samples, that demonstrate how to achieve certain functionality with enterprise Alexa skills, and in particular with Alexa for Business. Some samples are more complete, such as the Help Desk skill, but others will focus on specific components of a use case or integration.

Quivela2 is a tool to verify protocols modeled as object-oriented programs. Quivela is a prototype tool for constructing proofs of the security of cryptographic protocols.

The Alexa Auto SDK contains essential client-side software required to integrate Alexa into the automobile. The Auto SDK provides libraries that connect to Alexa and expose interfaces for your vehicle software to implement the platform-specific behavior for audio input, media streaming, calling through a connected phone, turn-by-turn navigation, controlling vehicle features such as heaters and lights, and

The Alexa APIs for Java consists of Java POJOS that represent the request and response JSON of Alexa services. These models act as a core dependency for the Alexa Skills Kit Java SDK These model classes are auto-generated using the JSON schemas in the developer documentation.