Applications of neural networks have expanded significantly in recent years from image segmentation to natural language processing to time-series forecasting. One notably successful use of deep learning is embedding, a method used to represent discrete variables as continuous vectors. This technique has found practical applications with word embeddings for machine translation and entity embeddings for categorical variables. In this article, I'll explain what neural network embeddings are, why we want to use them, and how they are learned. We'll go through these concepts in the context of a real problem I'm working on: representing all the books on Wikipedia as vectors to create a book recommendation system.

Deep learning (DL) has been the pillar of innovations in artificial intelligence. It has been providing a powerful way to compute numerous tough tasks. Now, many companies including Facebook and Google are leading DL research to strengthen their offerings. With the ever-growing research, DL is now venturing into the manufacturing process. This article highlights a recent study which used DL along with image processing to predict a laser's highest energy spot in a welding machine.

I was standing around at the Turing Institute in London recently and overheard someone say, "It's starting to smell a lot like algorithms in here." The Turing Institute is fragrant with algorithms. Part of the freshness of that smell is that deep neural networks, a kind of computing system inspired by the workings of the brain, are changing the way we think about learning. Some of their tricks are not that different from what psychologists have been describing for years. But that's actually good because the world through the eyes of the algorithm is just familiar enough that old ideas start to make sense in new ways.

Facebook may be willing to foot the bill. On Tuesday, Facebook and Udacity announced the PyTorch Scholarship Challenge, offering students the opportunity to learn how to build, train, and deploy deep learning models. PyTorch is an open source deep learning framework that is growing in popularity among AI researchers due to its ease of use, clean Pythonic API, and flexibility, Stuart Frye, Udacity's vice president of partnerships, wrote in a Tuesday blog post. With PyTorch 1.0, now available in preview release, developers can more easily move from exploration to product development with a single unified framework, Frye wrote. The scholarship program will offer students the chance to learn in-demand deep learning skills with PyTorch, as well as earn a full scholarship to Udacity's Deep Learning Nanodegree program, according to the post.

Artificial intelligence (AI) makes it possible for machines to learn from experience, adjust to new inputs and perform human-like tasks.

Meta-learning, or learning to learn, is the science of systematically observing how different machine learning approaches perform on a wide range of learning tasks, and then learning from this experience, or meta-data, to learn new tasks much faster than otherwise possible. Not only does this dramatically speed up and improve the design of machine learning pipelines or neural architectures, it also allows us to replace hand-engineered algorithms with novel approaches learned in a data-driven way. In this chapter, we provide an overview of the state of the art in this fascinating and continuously evolving field.

The recent practical successes [26] of Artificial Intelligence (AI) programs of the Reinforcement Learning and Deep Learning varieties in game playing, natural language processing and image classification, are now calling attention to the envisioned pitfalls of their hypothetical extension to wider domains of human behavior. Several voices from the industry and academia are now routinely raising concerns over the advances [49] of often heavily media-covered representatives of this new generation of programs such as Deep Blue, Watson, Google Translate, AlphaGo and AlphaZero. Most of these cutting-edge algorithms generally fall under the class of supervised learning, a branch of the still evolving taxonomy of Machine Learning techniques in AI research. In most cases the implementation choice is artificial neural networks software, the workhorse of the Connectionism school of thought in both AI and Cognitive Psychology. Confronting the current wave of connectionist architectures, critics usually raise issues of interpretability (Can the remarkable predictive capabilities be 1 trusted in real-life tasks? Are these capabilities transferable to unfamiliar situations or to different tasks altogether? How informative are the results about the real world; about human cognition?

Motion analysis is used in computer vision to understand the behaviour of moving objects in sequences of images. Optimising the interpretation of dynamic biological systems requires accurate and precise motion tracking as well as efficient representations of high-dimensional motion trajectories so that these can be used for prediction tasks. Here we use image sequences of the heart, acquired using cardiac magnetic resonance imaging, to create time-resolved three-dimensional segmentations using a fully convolutional network trained on anatomical shape priors. This dense motion model formed the input to a supervised denoising autoencoder (4Dsurvival), which is a hybrid network consisting of an autoencoder that learns a task-specific latent code representation trained on observed outcome data, yielding a latent representation optimised for survival prediction. To handle right-censored survival outcomes, our network used a Cox partial likelihood loss function. In a study of 302 patients the predictive accuracy (quantified by Harrell's C-index) was significantly higher (p < .0001) for our model C=0.73 (95$\%$ CI: 0.68 - 0.78) than the human benchmark of C=0.59 (95$\%$ CI: 0.53 - 0.65). This work demonstrates how a complex computer vision task using high-dimensional medical image data can efficiently predict human survival.

The boom of deep learning induced many industries and academies to introduce machine learning based approaches into their concern, competitively. However, existing machine learning frameworks are limited to sufficiently fulfill the collaboration and management for both data and models. We proposed NSML, a machine learning as a service (MLaaS) platform, to meet these demands. NSML helps machine learning work be easily launched on a NSML cluster and provides a collaborative environment which can afford development at enterprise scale. Finally, NSML users can deploy their own commercial services with NSML cluster. In addition, NSML furnishes convenient visualization tools which assist the users in analyzing their work. To verify the usefulness and accessibility of NSML, we performed some experiments with common examples. Furthermore, we examined the collaborative advantages of NSML through three competitions with real-world use cases.

Sensor fusion is a key technology that integrates various sensory inputs to allow for robust decision making in many applications such as autonomous driving and robot control. Deep neural networks have been adopted for sensor fusion in a body of recent studies. Among these, the so-called netgated architecture was proposed, which has demonstrated improved performances over the conventional convolutional neural networks (CNN). In this paper, we address several limitations of the baseline negated architecture by proposing two further optimized architectures: a coarser-grained gated architecture employing (feature) group-level fusion weights and a two-stage gated architectures leveraging both the group-level and feature level fusion weights. Using driving mode prediction and human activity recognition datasets, we demonstrate the significant performance improvements brought by the proposed gated architectures and also their robustness in the presence of sensor noise and failures.