Search results

Algorithms that fuse multiple input sources benefit from both complementary and shared information. Shared information may provide robustness against faulty or noisy inputs, which is indispensable for safety-critical applications like self-driving cars. We investigate learning fusion algorithms that are robust against noise added to a single source. We first demonstrate that robustness against single source

The Audio Front-End (AFE) is a key component in mitigating acoustic environmental challenges for far-field automatic speech recognition (ASR) on Amazon Echo family of products. A critical component of the AFE is the Beam Selector, which identifies which beam points to the target user. In this paper, we proposed a new SIR beam selector that utilizes subband-based signal-to-interference ratios to learn the

Automatic bandwidth extension (restoring high-frequency information from low sample rate audio) has a number of applications in speech processing. We introduce an end-to-end deep learning based system for speech bandwidth extension for use in a downstream automatic speech recognition (ASR) system. Specifically we propose a conditional generative adversarial network enriched with ASR-specific loss functions

We propose a novel approach for estimating the difficulty and transferability of supervised classification tasks. Unlike previous work, our approach is solution agnostic and does not require or assume trained models. Instead, we estimate these values using an information-theoretic approach: treating training labels as random variables and exploring their statistics. When transferring from a source to a target

Enabling cross-lingual NLP tasks by leveraging multilingual word embedding has recently attracted much attention. An important motivation is to support lower resourced languages, however, most efforts focus on demonstrating the effectiveness of the techniques using embeddings derived from similar languages toEnglish with large parallel content. In this study, we present a noise tolerant piecewise linear

In this paper, we propose a weak supervision framework for neural ranking tasks based on the data programming paradigm (Ratner et al., 2016), which enables us to leverage multiple weak supervision signals from different sources. Empirically, we consider two sources of weak supervision signals, unsupervised ranking functions and semantic feature similarities. We train a BERT-based passageranking model (which

Offline handwriting recognition with deep neural networks is usually limited to words or lines due to large computational costs. In this paper, a less computationally expensive full page offline handwritten text recognition framework is introduced. This framework includes a pipeline that locates handwritten text with an object detection neural network and recognises the text within the detected regions

This paper proposes a novel method to inject custom terminology into neural machine translation at run time. Previous works have mainly proposed modifications to the decoding algorithm in order to constrain the output to include run-time-provided target terms. While being effective, these constrained decoding methods add, however, significant computational overhead to the inference step, and, as we show

E-commerce companies like Amazon, Alibaba and Flipkart have an extensive catalogue comprising of billions of products. Matching customer search queries to plausible products is challenging due to the size and diversity of the catalogue. These challenges are compounded in apparel due to the semantic complexity and a large variation of fashion styles, product attributes and colours. Providing aids that can

In this age of social media, people often look at what others are wearing. In particular, Instagram and Twitter influencers often provide images of themselves wearing different outfits and their followers are often inspired to buy similar clothes.We propose a system to automatically find the closest visually similar clothes in the online Catalog (street-to-shop searching). The problem is challenging since

Transferring knowledge from a teacher neural network pretrained on the same or a similar task to a student neural network can significantly improve the performance of the student neural network. Existing knowledge transfer approaches match the activations or the corresponding handcrafted features of the teacher and the student networks. We propose an information-theoretic framework for knowledge transfer

We consider online forecasting problems for non-convex machine learning models. Forecasting introduces several challenges such as (i) frequent updates are necessary to deal with concept drift issues since the dynamics of the environment change overtime, and (ii) the state of the art models are non-convex models. We address these challenges with a novel regret framework. Standard regret measures commonly

Much of the recent progress made in image classification research can be credited to training procedure refinements, such as changes in data augmentations and optimization methods. In the literature, however, most refinements are either briefly mentioned as implementation details or only visible in source code. In this paper, we will examine a collection of such refinements and empirically evaluate their

We propose a method for specializing deep detectors and trackers to restricted settings. Our approach is designed with the following goals in mind: (a) Improving accuracy in restricted domains; (b) preventing overfitting to new domains and forgetting of generalized capabilities; (c) aggressive model compression and acceleration. To this end, we propose a novel loss that balances compression and acceleration

We introduce a new category of multivariate conditional generative models and demonstrate its performance and versatility in probabilistic time series forecasting and simulation. Specifically, the output of quantile regression networks is expanded from a set of fixed quantiles to the whole Quantile Function by a univariate mapping from a latent uniform distribution to the target distribution. Then the multivariate

We introduce a method to generate vectorial representations of visual classification tasks that can be used to reason about the nature of those tasks and their relations. Given a dataset with ground-truth labels and a loss function, we process images through a “probe network” and compute an embedding based on estimates of the Fisher information matrix associated with the probe network parameters. This provides

Entity linking is the task of linking mentions of named entities in natural language text, to entities in a curated knowledge-base. This is of significant importance in the biomedical domain, where it could be used to semantically annotate a large volume of clinical records and biomedical literature, to standardized concepts described in an ontology such as Unified Medical Language System (UMLS). We observe

This month, Amazon announced the 11 focus areas of the 2019 Amazon Research Awards.

We present an approach for expanding taxonomies with synonyms, or aliases. We target large shopping taxonomies, with thousands of nodes. A comprehensive set of entity aliases is an important component of identifying entities in unstructured text such as product reviews or search queries. Our method consists of two stages: we generate synonym candidates from WordNet and shopping search queries, then use

We introduce Gluon Time Series (GluonTS)1, a library for deep-learning-based time series modeling. GluonTS simplifies the development of and experimentation with time series models for common tasks such as forecasting or anomaly detection. It provides all necessary components and tools that scientists need for quickly building new models, for efficiently running and analyzing experiments and for evaluating