The power of machine learning systems not only promises great technical progress, but risks societal harm. As a recent example, researchers have shown that popular word embedding algorithms exhibit stereotypical biases, such as gender bias. The widespread use of these algorithms in machine learning systems, from automated translation services to curriculum vitae scanners, can amplify stereotypes in important contexts. Although methods have been developed to measure these biases and alter word embeddings to mitigate their biased representations, there is a lack of understanding in how word embedding bias depends on the training data. In this work, we develop a technique for understanding the origins of bias in word embeddings. Given a word embedding trained on a corpus, our method identifies how perturbing the corpus will affect the bias of the resulting embedding. This can be used to trace the origins of word embedding bias back to the original training documents. Using our method, one can investigate trends in the bias of the underlying corpus and identify subsets of documents whose removal would most reduce bias. We demonstrate our techniques on both a New York Times and Wikipedia corpus and find that our influence function-based approximations are extremely accurate.

Until now, using Google Translate on your iPhone has meant listening to the same pronunciation for translations no matter where you live. That's not very considerate, and potentially a problem if you live in countries where foreign accents could make comprehension difficult. You won't have that issue from now on -- an update to Google Translate has added speech output in local versions of multiple languages, including English, Bengali, French and Spanish. You can hear English results with an Indian accent, for instance, or listen to French with a Canadian spin. Android has included these speech options for a while.

We present Optimal Completion Distillation (OCD), a training procedure for optimizing sequence to sequence models based on edit distance. OCD is efficient, has no hyper-parameters of its own, and does not require pretraining or joint optimization with conditional log-likelihood. Given a partial sequence generated by the model, we first identify the set of optimal suffixes that minimize the total edit distance, using an efficient dynamic programming algorithm. Then, for each position of the generated sequence, we use a target distribution that puts equal probability on the first token of all the optimal suffixes. OCD achieves the state-of-the-art performance on end-to-end speech recognition, on both Wall Street Journal and Librispeech datasets, achieving $9.3\%$ WER and $4.5\%$ WER respectively.

It is not out of reason to boldly say that translation is of great importance to man. The diversity of languages and cultures in the world makes translation essential to humanity. The benefits of translation to humankind spread across businesses, politics, international relations, tourism, and education. Any company can go global. Moreover, the secret of a successful international business lies in quality translation services.

We present a sequence-to-action parsing approach for the natural language to SQL task that incrementally fills the slots of a SQL query with feasible actions from a pre-defined inventory. To account for the fact that typically there are multiple correct SQL queries with the same or very similar semantics, we draw inspiration from syntactic parsing techniques and propose to train our sequence-to-action models with non-deterministic oracles. We evaluate our models on the WikiSQL dataset and achieve an execution accuracy of 83.7% on the test set, a 2.1% absolute improvement over the models trained with traditional static oracles assuming a single correct target SQL query. When further combined with the execution-guided decoding strategy, our model sets a new state-of-the-art performance at an execution accuracy of 87.1%.

Music fans were the first to prove this by making a laughingstock of the app by loading lyrics from songs like Will Smith's "Fresh Prince of Bel-Air" and the theme song from Moana to see what funny or ridiculous translations Google would generate. While the tool isn't nearly as bad as videos make it out to be, this negative PR has kept companies from using it. After all, if Google can't translate song lyrics correctly, why would you trust it with marketing content? But Google Translate doesn't represent all machine translation. However, it is a brand that happens to be well-known and free.

Microsoft has brought its Translator to SwiftKey, allowing users to translate their conversations without having to leave the app they're in. With an update out today, SwiftKey for Android will translate incoming and outgoing messages in real time and it will be able to do so for over 60 languages. Additionally, while you won't need to install Microsoft Translator to be able to use the new SwiftKey feature, the company says translation will work offline if you do. Microsoft purchased SwiftKey in 2016 and it only makes sense that it would merge it's translator with the smart keyboard. Android users can access the feature through SwiftKey's Toolbar -- just tap the plus sign in the upper left corner of the keyboard to get there -- and you can check out which languages are supported here.

Last year in August Google released the Transformer, a novel neural network architecture based on a self-attention mechanism particularly well suited for language understanding. Before the Transformer, most neural network based approaches to machine translation relied on recurrent neural networks (RNNs) which operated sequentially using recurrence. In contrast to RNN-based approaches, the Transformer used no recurrence, instead it processed all words or symbols in the sequence and let each word attend the other word over multiple processing steps using a self-attention mechanism to incorporate context from words farther away. This approach led Transformer to train the recurrent models much faster and yield better translation results than RNNs. "However, on smaller and more structured language understanding tasks, or even simple algorithmic tasks such as copying a string (e.g. to transform an input of "abc" to "abcabc"), the Transformer does not perform very well.", says Stephan Gouws and Mostafa Dehghani from the Google Brain team. Hence this year the team has come up with Universal Transformers, an extension to standard Transformer which is computationally universal using a novel and efficient flavor of parallel-in-time recurrence.

It may not seem obvious, but your translation supply chain represents an enormous data security risk to your business. Particularly when you consider that hundreds of stakeholders could be involved in creating, translating, managing and delivering just one piece of content to customers. It's no surprise then, that security and data privacy have become boardroom issues. But recent advances in machine learning and artificial intelligence (AI) are changing everything, offering a fresh perspective and approach to some of the most difficult security challenges. Here we speak to Matthew Hardy, VP of Customer Solutions at SDL, about the complexities involved with managing a global content supply chain, and how AI brings new and exciting ways for brands to organize content in a way that improves security across all areas of their business.

Grammatical error correction, like other machine learning tasks, greatly benefits from large quantities of high quality training data, which is typically expensive to produce. While writing a program to automatically generate realistic grammatical errors would be difficult, one could learn the distribution of naturallyoccurring errors and attempt to introduce them into other datasets. Initial work on inducing errors in this way using statistical machine translation has shown promise; we investigate cheaply constructing synthetic samples, given a small corpus of human-annotated data, using an off-the-rack attentive sequence-to-sequence model and a straight-forward post-processing procedure. Our approach yields error-filled artificial data that helps a vanilla bi-directional LSTM to outperform the previous state of the art at grammatical error detection, and a previously introduced model to gain further improvements of over 5% $F_{0.5}$ score. When attempting to determine if a given sentence is synthetic, a human annotator at best achieves 39.39 $F_1$ score, indicating that our model generates mostly human-like instances.