The BERT model has arisen as a popular state-of-the-art machine learning model in the recent years that is able to cope with multiple NLP tasks such as supervised text classification without human supervision. Its flexibility to cope with any type of corpus delivering great results has make this approach very popular not only in academia but also in the industry. Although, there are lots of different approaches that have been used throughout the years with success. In this work, we first present BERT and include a little review on classical NLP approaches. Then, we empirically test with a suite of experiments dealing different scenarios the behaviour of BERT against the traditional TF-IDF vocabulary fed to machine learning algorithms. Our purpose of this work is to add empirical evidence to support or refuse the use of BERT as a default on NLP tasks. Experiments show the superiority of BERT and its independence of features of the NLP problem such as the language of the text adding empirical evidence to use BERT as a default technique to be used in NLP problems.

The language barrier is preventing many foreign residents in Hiroshima Prefecture from keeping abreast of the latest status of the coronavirus pandemic, highlighting the need for municipalities to provide essential information in multiple languages. "It was through social media that I came to know about the whole kyūgyō yōsei thing," Michelle Crothers, an Australian who runs an English conversation school in the city of Hiroshima, said, referring to the Japanese phrase for "request to suspend businesses." On April 18, when the prefecture issued the request in line with the state of emergency declared by the central government, Crothers stumbled upon a friend's social media post written in English about the prefecture's announcement. She then fumbled her way through official websites by the government and the prefecture in hopes of finding out whether her school will have to shut down in line with the request, but ended up giving up. As an extra precaution, she decided to close it for the time being.

Dual learning has been successfully applied in many machine learning applications including machine translation, image-to-image transformation, etc. The high-level idea of dual learning is very intuitive: if we map an $x$ from one domain to another and then map it back, we should recover the original $x$. Although its effectiveness has been empirically verified, theoretical understanding of dual learning is still very limited. In this paper, we aim at understanding why and when dual learning works. Based on our theoretical analysis, we further extend dual learning by introducing more related mappings and propose multi-step dual learning, in which we leverage feedback signals from additional domains to improve the qualities of the mappings. We prove that multi-step dual learn-ing can boost the performance of standard dual learning under mild conditions. Experiments on WMT 14 English$\leftrightarrow$German and MultiUNEnglish$\leftrightarrow$French translations verify our theoretical findings on dual learning, and the results on the translations among English, French, and Spanish of MultiUN demonstrate the effectiveness of multi-step dual learning.

The task of transliteration of named entities from one language into another is complicated and considered as one of the challenging tasks in machine translation (MT). To build a well performed transliteration system, we apply well-established techniques based on Hybrid Deep Learning. The system based on convolutional neural network (CNN) followed by Bi-LSTM and CRF. The proposed hybrid mechanism is examined on ANERCorp and Kalimat corpus. The results show that the neural machine translation approach can be employed to build efficient machine transliteration systems achieving state-of-the-art results for Arabic – English language.

Neural Machine Translation (NMT) relies heavily on word embeddings, which are continuous representations of words in a vector space, obtained from large monolingual data and, independently, from bilingual data for NMT model training. Word embeddings have proven to be invaluable for performance improvements in natural language analysis tasks that otherwise suffer from data scarcity. This paper defines a new cost function---demonstrated on Farsi-Spanish low-resource attention-based NMT---that encodes word similarity as distances within a word embedding space. The novelty of this cost function is that it encourages our attentional NMT model to generate words that are close to their references in the embedding space. This approach encourages the decoder to select acceptably similar words when potential candidates are found to be Out-Of-Vocabulary (OOV). Experimental results demonstrate improvements of our attentional NMT model over a community-standard NMT baseline model.

Training on only perfect Standard English corpora predisposes pre-trained neural networks to discriminate against minorities from non-standard linguistic backgrounds (e.g., African American Vernacular English, Colloquial Singapore English, etc.). We perturb the inflectional morphology of words to craft plausible and semantically similar adversarial examples that expose these biases in popular NLP models, e.g., BERT and Transformer, and show that adversarially fine-tuning them for a single epoch significantly improves robustness without sacrificing performance on clean data.

The choice of hyper-parameters affects the performance of neural models. While much previous research (Sutskever et al., 2013; Duchi et al., 2011; Kingma and Ba, 2015) focuses on accelerating convergence and reducing the effects of the learning rate, comparatively few papers concentrate on the effect of batch size. In this paper, we analyze how increasing batch size affects gradient direction, and propose to evaluate the stability of gradients with their angle change. Based on our observations, the angle change of gradient direction first tends to stabilize (i.e. gradually decrease) while accumulating mini-batches, and then starts to fluctuate. We propose to automatically and dynamically determine batch sizes by accumulating gradients of mini-batches and performing an optimization step at just the time when the direction of gradients starts to fluctuate. To improve the efficiency of our approach for large models, we propose a sampling approach to select gradients of parameters sensitive to the batch size. Our approach dynamically determines proper and efficient batch sizes during training. In our experiments on the WMT 14 English to German and English to French tasks, our approach improves the Transformer with a fixed 25k batch size by +0.73 and +0.82 BLEU respectively.

In work implicitly includes monolingual and natural language processing, the main promise of cross-lingual signals that constitute a departure multilingual learning is to bridge the digital language from the pure setting. We review existing training divide, to enable access to information and signals as well as other signals that may be technology for the world's 6,900 languages (Ruder of interest for future study (§4). We then discuss et al., 2019). For the purpose of this paper, we methodological issues in UCL (e.g., validation, hyperparameter define "multilingual learning" as learning a common tuning) and propose best evaluation model for two or more languages from raw practices (§5). Finally, we provide a unified outlook text, without any downstream task labels. Common of established research areas (cross-lingual use cases include translation as well as pretraining word embeddings, deep multilingual models and multilingual representations. We will use the term unsupervised machine translation) in UCL (§6), interchangeably with "cross-lingual learning".

Neural networks are known to suffer from catastrophic forgetting when trained on sequential datasets. While there have been numerous attempts to solve this problem for large-scale supervised classification, little has been done to overcome catastrophic forgetting for few-shot classification problems. We demonstrate that the popular gradient-based few-shot meta-learning algorithm Model-Agnostic Meta-Learning (MAML) indeed suffers from catastrophic forgetting and introduce a Bayesian online meta-learning framework that tackles this problem. Our framework incorporates MAML into a Bayesian online learning algorithm with Laplace approximation. This framework enables few-shot classification on a range of sequentially arriving datasets with a single meta-learned model. The experimental evaluations demonstrate that our framework can effectively prevent forgetting in various few-shot classification settings compared to applying MAML sequentially.

In data-parallel synchronous training of deep neural networks, different devices (replicas) run the same program with different partitions of the training batch, but weight update computation is repeated on all replicas, because the weights do not have a batch dimension to partition. This can be a bottleneck for performance and scalability in typical language models with large weights, and models with small per-replica batch size which is typical in large-scale training. This paper presents an approach to automatically shard the weight update computation across replicas with efficient communication primitives and data formatting, using static analysis and transformations on the training computation graph. We show this technique achieves substantial speedups on typical image and language models on Cloud TPUs, requiring no change to model code. This technique helps close the gap between traditionally expensive (ADAM) and cheap (SGD) optimizers, as they will only take a small part of training step time and have similar peak memory usage. It helped us to achieve state-of-the-art training performance in Google's MLPerf 0.6 submission.