OpenAI stunned the world with the release of Generative Pre-trained Transformer 3 (GPT-3), the world's most impressive language-generating AI. OpenAI CEO Sam Altman joins Azeem Azhar to reflect on the huge attention generated by GPT-3 and what it heralds for the future research and development toward the creation of a true artificial general intelligence (AGI). HBR Presents is a network of podcasts curated by HBR editors, bringing you the best business ideas from the leading minds in management. The views and opinions expressed are solely those of the authors and do not necessarily reflect the official policy or position of Harvard Business Review or its affiliates.

We release unofficial pytorch code for "FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence,", which accepted in NeurIPS'20!! I found that some pytorch implementations are already released, but often fail to reproduce the results in the paper. We hope our implementation helps SSL researchers make their awesome projects! If you like our project, please use and make many issues, and mark the star!

Optimizers are learning algorithm which used to train our Neural Networks. This is one of the most important part in our model.Continue reading on Medium » Source

What did we learn implementing GPT-2 and BERT for Conversational AI in Healthcare?Continue reading on Inception Health » Source

Machine-learning algorithms use statistics to find patterns in massive* amounts of data. And data, here, encompasses a lot of things--numbers, words, images, clicks, what have you. If it can be digitally stored, it can be fed into a machine-learning algorithm. Machine learning is the process that powers many of the services we use today--recommendation systems like those on Netflix, YouTube, and Spotify; search engines like Google and Baidu; social-media feeds like Facebook and Twitter; voice assistants like Siri and Alexa.

The MNIST dataset is the most overused dataset for getting started with image classification. MNIST dataset comprising of 10-class handwritten digits introduced by Yann LeCun in 1998 come up over and over again, in scientific papers, blog posts, and so on. It contains 28 28 (also 32x32) grayscale images of handwritten digits, each with integers between 0 and 9. The reason MNIST is so popular has to do with its size, allowing deep learning practitioners to quickly check, train, and publish their algorithms. There are certain variations and limitations of MNIST.

In this work we propose for the first time a transformer-based framework for unsupervised representation learning of multivariate time series. Pre-trained models can be potentially used for downstream tasks such as regression and classification, forecasting and missing value imputation. By evaluating our models on several benchmark datasets for multivariate time series regression and classification, we show that not only does our modeling approach represent the most successful method employing unsupervised learning of multivariate time series presented to date, but also that it exceeds the current state-of-the-art performance of supervised methods; it does so even when the number of training samples is very limited, while offering computational efficiency. Finally, we demonstrate that unsupervised pre-training of our transformer models offers a substantial performance benefit over fully supervised learning, even without leveraging additional unlabeled data, i.e., by reusing the same data samples through the unsupervised objective.

Automatic prediction of age and gender from face images has drawn a lot of attention recently, due it is wide applications in various facial analysis problems. However, due to the large intra-class variation of face images (such as variation in lighting, pose, scale, occlusion), the existing models are still behind the desired accuracy level, which is necessary for the use of these models in real-world applications. In this work, we propose a deep learning framework, based on the ensemble of attentional and residual convolutional networks, to predict gender and age group of facial images with high accuracy rate. Using attention mechanism enables our model to focus on the important and informative parts of the face, which can help it to make a more accurate prediction. We train our model in a multi-task learning fashion, and augment the feature embedding of the age classifier, with the predicted gender, and show that doing so can further increase the accuracy of age prediction. Our model is trained on a popular face age and gender dataset, and achieved promising results. Through visualization of the attention maps of the train model, we show that our model has learned to become sensitive to the right regions of the face.

Natural language descriptions of user interface (UI) elements such as alternative text are crucial for accessibility and language-based interaction in general. Yet, these descriptions are constantly missing in mobile UIs. We propose widget captioning, a novel task for automatically generating language descriptions for UI elements from multimodal input including both the image and the structural representations of user interfaces. We collected a large-scale dataset for widget captioning with crowdsourcing. Our dataset contains 162,859 language phrases created by human workers for annotating 61,285 UI elements across 21,750 unique UI screens. We thoroughly analyze the dataset, and train and evaluate a set of deep model configurations to investigate how each feature modality as well as the choice of learning strategies impact the quality of predicted captions. The task formulation and the dataset as well as our benchmark models contribute a solid basis for this novel multimodal captioning task that connects language and user interfaces.

In this study, the effects of different class labels created as a result of multiple conceptual meanings on localization using Weakly Supervised Learning presented on Car Dataset. In addition, the generated labels are included in the comparison, and the solution turned into Unsupervised Learning. This paper investigates multiple setups for car localization in the images with other approaches rather than Supervised Learning. To predict localization labels, Class Activation Mapping (CAM) is implemented and from the results, the bounding boxes are extracted by using morphological edge detection. Besides the original class labels, generated class labels also employed to train CAM on which turn to a solution to Unsupervised Learning example. In the experiments, we first analyze the effects of class labels in Weakly Supervised localization on the Compcars dataset. We then show that the proposed Unsupervised approach outperforms the Weakly Supervised method in this particular dataset by approximately %6.