DeepMind has successfully developed a system that can analyze retinal scans and spot symptoms of sight-threatening eye diseases. Today, the AI division -- owned by Google's parent company Alphabet -- published "early results" of a research project with the UK's Moorfields Eye Hospital. They show that the company's algorithms can quickly examine optical coherence tomography (OCT) scans and make diagnoses with the same accuracy as human clinicians. In addition, the system can show its workings, allowing eye care professionals to scrutinize the final assessment. At the moment, hospitals and clinics use flesh-and-bone specialists to dissect OCT scans.

A small team of student AI (artificial intelligence) coders outperformed codes from Google's researchers, reveal an important benchmark. Students from Fast.ai, a non-profit group that creates learning resources and is dedicated to making deep learning "accessible to all", have created an AI algorithm that beats code from Google's researchers. Researchers from Stanford measured the algorithm using a benchmark called DAWNBench that uses a common image classification task to track the speed of a deep-learning algorithm per dollar of compute power. According to the benchmark, the researchers found that the algorithm built by Fast.ai's team had beaten Google's code. Fast.ai consists of part-time students who are eager to try out machine learning and convert it into a career in data science.

Neural Networks are a class of models within the general machine learning literature. So for example, if you took a Coursera course on machine learning, neural networks will likely be covered. Neural networks are a specific set of algorithms that has revolutionized the field of machine learning. They are inspired by biological neural networks and the current so called deep neural networks have proven to work quite very well. Neural Networks are themselves general function approximations, that is why they can be applied to literally almost any machine learning problem where the problem is about learning a complex mapping from the input to the output space. After finishing the famous Andrew Ng's Machine Learning Coursera course, I started developing interest towards neural networks and deep learning.

"OpenAI Five plays 180 years worth of games against itself every day." Artificial Intelligence vs Humans is a hotly debated topic these days and the gaming arena holds one of the prime witnesses of that. DOTA 2 players have taken on an AI algorithm before in a head-to-head match and have lost. The story seems to continue to date, a recent match between a team of 5 humans vs a mix of AI programs, being a proof of that. The AI programs, developed by OpenAI - an AI research lab founded by Elon Musk and Y Combinator president Sam Altman, won 2 out of 3 matches against the 5 semi-professional humans working together as a team.

The way that AI is talked about today, you could be forgiven for thinking that it will'just happen'. It sometimes sounds as if machine learning is on an unstoppable roll, poised to revolutionise industry and commerce through sheer momentum of technology. But of course, it's not that simple for organisations that want to adopt AI methods. Indeed, the technologies are maturing at an unprecedented rate. There are countless innovations already being developed to help diagnose deadly diseases, make million-dollar trading decisions, learn skills like driving, and automate many other tasks that have the power to transform entire industries. But of course we know that AI is not magic.

By Rachel Thomas, Co-founder at fast.ai Editor's note: This is part 3 in a series. To announce Google's AutoML, Google CEO Sundar Pichai wrote, "Today, designing neural nets is extremely time intensive, and requires an expertise that limits its use to a smaller community of scientists and engineers. That's why we've created an approach called AutoML, showing that it's possible for neural nets to design neural nets. We hope AutoML will take an ability that a few PhDs have today and will make it possible in three to five years for hundreds of thousands of developers to design new neural nets for their particular needs."

Summarizing, there are four desiderata that one would like to see in A.I. systems, if they have to compare well with human (or just animal) brains: meta-learning, learning by demonstration ("few-shot learning"), transfer learning and multi-task learning. Meta-learning is particularly relevant in the case of reinforcement learning. It is obvious that reinforcement learning is highly unnatural. DeepMind's AlphaGo and OpenAi Five need to learn from scratch via a huge number of trials. Animals, instead, use built-in or acquired "meta-skills" to learn new tasks in just a few trials. Modern computational theory of meta-learning (learning how to learn) dates back at least to the 1990s, when Schmidhuber published the manifesto "Simple Principles of Metalearning" (1996), followed by his student Sepp Hochreiter ("Learning to Learn Using Gradient Descent", 2001), and by Nicolas Schweighofer and Kenji Doya at Japan's ATR ("Meta-learning in Reinforcement Learning", 2001). Examples of "deep" meta-learning systems of the new generation are: RL Square by Pieter Abbeel's student Yan Duan at UC Berkeley, based on Schulman's TRPO ("RL Square: Fast Reinforcement Learning via Slow Reinforcement Learning", 2016); the "model-agnostic meta-learning" (MAML) of Sergey Levine's student Chelsea Finn at UC Berkeley ("Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks", 2017); Marcel Binz's thesis at KTH Royal Institute of Technology ("Learning Goal-Directed Behaviour", 2017); Jane Wang's "deep meta-reinforcement learning" at DeepMind ("Learning to Reinforcement Learn", 2017); and OpenAI's Reptile, developed by Alex Nichol and John Schulman, a generalization of Finn's MAML ("On First-Order Meta-Learning Algorithms", 2018). DeepMind's neuroscientist Matthew Botvinick believes that the latter could be a model for how our brain learns: the dopamine system trains another part of the brain, the prefrontal cortex, to operate as its own free-standing learning system ("Prefrontal Cortex as a Meta-reinforcement Learning System", 2018).

We present hidden fluid mechanics (HFM), a physics informed deep learning framework capable of encoding an important class of physical laws governing fluid motions, namely the Navier-Stokes equations. In particular, we seek to leverage the underlying conservation laws (i.e., for mass, momentum, and energy) to infer hidden quantities of interest such as velocity and pressure fields merely from spatio-temporal visualizations of a passive scaler (e.g., dye or smoke), transported in arbitrarily complex domains (e.g., in human arteries or brain aneurysms). Our approach towards solving the aforementioned data assimilation problem is unique as we design an algorithm that is agnostic to the geometry or the initial and boundary conditions. This makes HFM highly flexible in choosing the spatio-temporal domain of interest for data acquisition as well as subsequent training and predictions. Consequently, the predictions made by HFM are among those cases where a pure machine learning strategy or a mere scientific computing approach simply cannot reproduce. The proposed algorithm achieves accurate predictions of the pressure and velocity fields in both two and three dimensional flows for several benchmark problems motivated by real-world applications. Our results demonstrate that this relatively simple methodology can be used in physical and biomedical problems to extract valuable quantitative information (e.g., lift and drag forces or wall shear stresses in arteries) for which direct measurements may not be possible.

In this paper, we will show an unprecedented method to accelerate training and improve performance, which called random gradient (RG). This method can be easier to the training of any model without extra calculation cost, we use Image classification, Semantic segmentation, and GANs to confirm this method can improve speed which is training model in computer vision. The central idea is using the loss multiplied by a random number to random reduce the back-propagation gradient. We can use this method to produce a better result in Pascal VOC, Cifar, Cityscapes datasets.

Background: It is still an open research area to theoretically understand why Deep Neural Networks (DNNs)-- equipped with many more parameters than training data and trained by (stochastic) gradient-based methods-- often achieve remarkably low generalization error. Contribution: We study DNN training by Fourier analysis. Our theoretical framework explains: i) DNN with (stochastic) gradient-based methods endows low-frequency components of the target function with a higher priority during the training; ii) Small initialization leads to good generalization ability of DNN while preserving the DNN's ability of fitting any function. These results are further confirmed by experiments of DNNs fitting the following datasets, i.e., natural images, one-dimensional functions and MNIST dataset.