Search results

Dialogue response generation models that use template ranking rather than direct sequence generation allow model developers to limit generated responses to pre-approved messages. However, manually creating templates is time consuming and requires domain expertise. To alleviate this problem, we explore automating the process of creating dialogue templates by using unsupervised methods to cluster historical

Electronic records contain sequences of events, some of which take place all at once in a single visit, and others that are dispersed over multiple visits, each with a different timestamp. We postulate that fine temporal detail, e.g., whether a series of blood tests are completed at once or in rapid succession should not alter predictions based on this data. Motivated by this intuition, we propose models

Nowadays, as cameras are rapidly adopted in our daily routine, images of documents are becoming both abundant and prevalent. Unlike natural images that capture physical objects, document-images contain a significant amount of text with critical semantics and complicated layouts. In this work, we devise a generic unsupervised technique to learn multimodal affinities between textual entities in a document-image

In object-oriented languages, constructors often have a combination of required and optional formal parameters. It is tedious and inconvenient for programmers to write a constructor by hand for each combination. The multitude of constructors is error-prone for clients, and client code is difficult to read due to the large number of constructor arguments. Therefore, programmers often use design patterns

The recent emergence of multilingual pretraining language model (mPLM) has enabled breakthroughs on various downstream crosslingual transfer (CLT) tasks. However, mPLMbased methods usually involve two problems: (1) simply fine-tuning may not adapt generalpurpose multilingual representations to be task-aware on low-resource languages; (2) ignore how cross-lingual adaptation happens for downstream tasks.

Conversational agents are consistently growing in popularity and many people interact with them every day. While many conversational agents act as personal assistants, they can have many different goals. Some are task-oriented, such as providing customer support for a bank or making a reservation. Others are designed to be empathetic and to form emotional connections with the user. The Alexa Prize Challenge

We show that a variant of the surface code — the XZZX code — offers remarkable performance for fault-tolerant quantum computation. The error threshold of this code achieves the zero-rate hashing bound for every single-qubit Pauli noise channel; it is the first explicit code shown to have this universal property. We present numerical evidence that this threshold even exceeds the hashing bound for an experimentally

An invaluable method for probing the physics of a quantum many-body spin system is a mapping to non-interacting effective fermions. We find such mappings using only the frustration graph G of a Hamiltonian H, i.e., the network of anti-commutation relations between the Pauli terms in H in a given basis. We prove theorems based solely on the graph G to identify when H is a free-fermion-solvable model, even

Precipitation nowcasting refers to the forecasting of rainfall and other types of precipitation up to 6 hours ahead (as defined by the World Meteorological Organization)1. Since rainfall can be localized and highly changeable, users of precipitation nowcast typically demand to know the exact time, location and intensity of rainfall. It is therefore necessary to make very high resolution, both spatially

TLS 1.3 allows two parties to establish a shared session key from an out-of-band agreed Pre Shared Key (PSK). The PSK is used to mutually authenticate the parties, under the assumption that it is not shared with others. This allows the parties to skip the certificate verification steps, saving bandwidth, communication rounds, and latency. We identify a security vulnerability in this TLS 1.3 path, by showing

We discuss promising recent contributions on quantifying feature relevance using Shapley values, where we observed some confusion on which probability distribution is the right one for dropped features. We argue that the confusion is based on not carefully distinguishing between observational and interventional conditional probabilities and try a clarification based on Pearl’s seminal work on causality.

When customers visit an ecommerce website, they will perform certain actions and will eventually either make a purchase or end their session without a purchase. Website operators can use the browsing behavior of their customers to build machine learning models that allow them to target customers that are more likely to convert with promotions. In this solution we will demonstrate how one can use SageMaker

Pretrained masked language models (MLMs) require finetuning for most NLP tasks. Instead, we evaluate MLMs out of the box via their pseudo-log-likelihood scores (PLLs), which are computed by masking tokens one by one. We show that PLLs outperform scores from autoregressive language models like GPT-2 in a variety of tasks. By rescoring ASR and NMT hypotheses, RoBERTa reduces an end-to-end LibriSpeech model

The Alexa Skills Kit SDK for Java helps you get a skill up and running quickly, letting you focus on skill logic instead of boilerplate code.

Amazon SageMaker Debugger is designed to be a debugger for machine learning models. It lets you go beyond just looking at scalars like losses and accuracies during training and gives you full visibility into all tensors 'flowing through the graph' during training or inference. Amazon SageMaker Debugger RulesConfig provides a mapping of builtin rules with default configurations. These configurations will

While models have reached superhuman performance on popular question answering (QA) datasets such as SQuAD, they have yet to outperform humans on the task of question answering itself. In this paper, we investigate if models are learning reading comprehension from QA datasets by evaluating BERT-based models across five datasets. We evaluate models on their generalizability to out-of-domain examples, responses

This project shows how to use Deep Graph Library (DGL) on Amazon SageMaker to train a graph neural network (GNN) model to perform entity resolution on customer identity graphs. See the project detail page to learn more about the techniques used.

Knowledge graphs have emerged as a key abstraction for organizing information in diverse domains and their embeddings are increasingly used to harness their information in various information retrieval and machine learning tasks. However, the ever growing size of knowledge graphs requires computationally efficient algorithms capable of scaling to graphs with millions of nodes and billions of edges. This

This paper introduces Meta-Q-Learning (MQL), a new off-policy algorithm for meta-Reinforcement Learning (meta-RL). MQL builds upon three simple ideas. First, we show that Q-learning is competitive with state-of-the-art meta-RL algorithms if given access to a context variable that is a representation of the past trajectory. Second, a multi-task objective to maximize the average reward across the training

The Well-Architected framework has been developed to help cloud architects build the most secure, high-performing, resilient, and efficient infrastructure possible for their applications. This framework provides a consistent approach for customers and partners to evaluate architectures, and provides guidance to help implement designs that will scale with your application needs over time. This repository