Large, pre-trained transformer-based language models such as BERT have drastically changed the Natural Language Processing (NLP) field. We present a survey of recent work that uses these large language models to solve NLP tasks via pre-training then fine-tuning, prompting, or text generation approaches. We also present approaches that use pre-trained language models to generate data for training augmentation or other purposes. We conclude with discussions on limitations and suggested directions for future research.

The largest dataset of Arabic speech mispronunciation detections in Egyptian dialogues is introduced. The dataset is composed of annotated audio files representing the top 100 words that are most frequently used in the Arabic language, pronounced by 100 Egyptian children (aged between 2 and 8 years old). The dataset is collected and annotated on segmental pronunciation error detections by expert listeners.

This paper reports on the evaluation of Deep Learning (DL) transformer architecture models for Named-Entity Recognition (NER) on ten low-resourced South African (SA) languages. In addition, these DL transformer models were compared to other Neural Network and Machine Learning (ML) NER models. The findings show that transformer models significantly improve performance when applying discrete fine-tuning parameters per language. Furthermore, fine-tuned transformer models outperform other neural network and machine learning models with NER on the low-resourced SA languages. For example, the transformer models generated the highest F-scores for six of the ten SA languages, including the highest average F-score surpassing the Conditional Random Fields ML model. Additional research could evaluate the more recent transformer architecture models on other Natural Language Processing tasks and applications, such as Phrase chunking, Machine Translation, and Part-of-Speech tagging.

Artificial intelligence, also known as AI, is intelligence demonstrated by machines, as opposed to the natural intelligence displayed by humans. AI applications include advanced web search engines, recommendation systems, speech recognition, self-driving cars, and much more. Now AI is involved in the areas of writing, both fiction and non-fiction, and graphics. First, let's start out with art. The woman that you see above was created with artificial intelligence (and a little input from me).

The popularity of machine learning and artificial intelligence is driving more and more technological innovations. The tech market is also attracting several new tech professionals, both from tech and non-tech backgrounds. The emergence of machine learning hackathons has turned out to be one of the best ways for machine learning and AI practitioners to practice and show off their skills. Hackathons provide an environment for the participants to work on various kinds of projects using distinct tools to show off their skills. In this article, we talk about the top machine learning hackathons that AI professionals can choose from in 2021.

Over the past several decades, climate change has led to major disasters in Eastern Africa countries including Tanzania. From floods, chronic droughts, landslides, strong winds and earthquakes to their secondary impacts of diseases and epidemics, these are some of the recent disasters plaguing Tanzania. These disasters lead to death and displacement of people, loss of properties and livelihoods, disruption of social networks and services such as water, food, and healthcare thereby leaving communities more vulnerable and susceptible to the next extreme event. Lack of disaster preparedness and awareness makes the situation worse as communities remain helpless in the event of disasters hence face its full impact. Combining citizen science and modern technological innovation provides an opportunity to build the resilience of communities and reduce risks.

For Bayesian optimization (BO) on high-dimensional data with complex structure, neural network-based kernels for Gaussian processes (GPs) have been used to learn flexible surrogate functions by the high representation power of deep learning. However, existing methods train neural networks by maximizing the marginal likelihood, which do not directly improve the BO performance. In this paper, we propose a meta-learning method for BO with neural network-based kernels that minimizes the expected gap between the true optimum value and the best value found by BO. We model a policy, which takes the current evaluated data points as input and outputs the next data point to be evaluated, by a neural network, where neural network-based kernels, GPs, and mutual information-based acquisition functions are used as its layers. With our model, the neural network-based kernel is trained to be appropriate for the acquisition function by backpropagating the gap through the acquisition function and GP. Our model is trained by a reinforcement learning framework from multiple tasks. Since the neural network is shared across different tasks, we can gather knowledge on BO from multiple training tasks, and use the knowledge for unseen test tasks. In experiments using three text document datasets, we demonstrate that the proposed method achieves better BO performance than the existing methods.

The summit will bring together TensorFlow and machine learning enthusiast in SSA for a two day event that will feature talks on new developments in TensorFlow, Machine Learning in the browser, as on-device Machine Learning at the edge. You will hear from the TensorFlow team and Machine Learning Google Developer Experts across Africa and the rest of the world. You will hear from Machine Learning Google Developer Experts on latest updates!. Do register to get update and link to join the call when we are live. English Session will be streamed from and recording will be at https://bit.ly/tfug-ssa-live

As the Internet grows in size, so does the amount of text based information that exists. For many application spaces it is paramount to isolate and identify texts that relate to a particular topic. While one-class classification would be ideal for such analysis, there is a relative lack of research regarding efficient approaches with high predictive power. By noting that the range of documents we wish to identify can be represented as positive linear combinations of the Vector Space Model representing our text, we propose Conical classification, an approach that allows us to identify if a document is of a particular topic in a computationally efficient manner. We also propose Normal Exclusion, a modified version of Bi-Normal Separation that makes it more suitable within the one-class classification context. We show in our analysis that our approach not only has higher predictive power on our datasets, but is also faster to compute.

Fast and accurate prediction of hurricane evolution from genesis onwards is needed to reduce loss of life and enhance community resilience. In this work, a novel model development methodology for predicting storm trajectory is proposed based on two classes of Recurrent Neural Networks (RNNs). The RNN models are trained on input features available in or derived from the HURDAT2 North Atlantic hurricane database maintained by the National Hurricane Center (NHC). The models use probabilities of storms passing through any location, computed from historical data. A detailed analysis of model forecasting error shows that Many-To-One prediction models are less accurate than Many-To-Many models owing to compounded error accumulation, with the exception of $6-hr$ predictions, for which the two types of model perform comparably. Application to 75 or more test storms in the North Atlantic basin showed that, for short-term forecasting up to 12 hours, the Many-to-Many RNN storm trajectory prediction models presented herein are significantly faster than ensemble models used by the NHC, while leading to errors of comparable magnitude.