Google has been tinkering in the field of medicine over the last few years, including developing a prototype electronic contact lens. The company's latest health project involves detecting cardiovascular conditions by analyzing the vasculature of the retina. The researchers built a deep learning system that processed data from two datasets containing thousands of patients, each of which included images of a patient's retina along with various risk factors and health conditions such smoking and high blood pressure. The system found correlations between various parameters measured within the retinal images and cardiovascular risk factors, as well as disease. For example, the software was able to identify smokers just by looking at the retina 71% of the time.

A standard introduction to online learning might place Online Gradient Descent at its center and then proceed to develop generalizations and extensions like Online Mirror Descent and secondorder methods. Here we explore the alternative approach of putting exponential weights (EW) first. We show that many standard methods and their regret bounds then follow as a special case by plugging in suitable surrogate losses and playing the EW posterior mean. For instance, we easily recover Online Gradient Descent by using EW with a Gaussian prior on linearized losses, and, more generally, all instances of Online Mirror Descent based on regular Bregman divergences also correspond to EW with a prior that depends on the mirror map. Furthermore, appropriate quadratic surrogate losses naturally give rise to Online Gradient Descent for strongly convex losses and to Online Newton Step. We further interpret several recent adaptive methods (iProd, Squint, and a variation of Coin Betting for experts) as a series of closely related reductions to exp-concave surrogate losses that are then handled by Exponential Weights. Finally, a benefit of our EW interpretation is that it opens up the possibility of sampling from the EW posterior distribution instead of playing the mean. As already observed by Bubeck and Eldan (2015), this recovers the best-known rate in Online Bandit Linear Optimization.

Humans and animals have the ability to continually acquire and fine-tune knowledge throughout their lifespan. This ability is mediated by a rich set of neurocognitive functions that together contribute to the early development and experience-driven specialization of our sensorimotor skills. Consequently, the ability to learn from continuous streams of information is crucial for computational learning systems and autonomous agents (inter)acting in the real world. However, continual lifelong learning remains a long-standing challenge for machine learning and neural network models since the incremental acquisition of new skills from non-stationary data distributions generally leads to catastrophic forgetting or interference. This limitation represents a major drawback also for state-of-the-art deep neural network models that typically learn representations from stationary batches of training data, thus without accounting for situations in which the number of tasks is not known a priori and the information becomes incrementally available over time. In this review, we critically summarize the main challenges linked to continual lifelong learning for artificial learning systems and compare existing neural network approaches that alleviate, to different extents, catastrophic interference. Although significant advances have been made in domain-specific continual lifelong learning with neural networks, extensive research efforts are required for the development of general-purpose artificial intelligence and autonomous agents. We discuss well-established research and recent methodological trends motivated by experimentally observed lifelong learning factors in biological systems. Such factors include principles of neurosynaptic stability-plasticity, critical developmental stages, intrinsically motivated exploration, transfer learning, and crossmodal integration.

Did you also see the video of the robot opening a door? Did you freak out and think: if they can open doors…what will they learn next? People see a robot do something that looks intelligent and the natural instinct is to extrapolate and assume that there is a general intelligence behind the action. In reality, the set of tasks that robots and Artificial Intelligence (AI) more generally are good at right now is constrained by very narrow parameters. Think about the game of chess (or more recently, the Chinese game of Go).

In this course we are going to look at advanced NLP. Previously, you learned about some of the basics, like how many NLP problems are just regular machine learning and data science problems in disguise, and simple, practical methods like bag-of-words and term-document matrices. These allowed us to do some pretty cool things, like detect spam emails, write poetry, spin articles, and group together similar words. In this course I'm going to show you how to do even more awesome things. We'll learn not just 1, but 4 new architectures in this course.

Emerging technologies like machine learning, deep learning, and, to some extent, artificial intelligence are becoming ripe for adoption. IT professionals need to understand how these technologies affect the businesses they support and begin developing the skills needed to manage and work with these technologies as they evolve. It's only a matter of time before this emerging technology becomes not only commonplace but also essential for IT pros to successfully do their jobs. According to a new Gartner report, by 2022 one in five workers engaged in non-routine tasks will rely on AI to help with their work. The benefits of the analytics and predictive capabilities enabled by ML and AI are far-reaching.

Organizations expend constant effort to deliver information employees need to know for their jobs. You depend on training to help your employees make more sales, provide better customer service, avoid regulatory issues, and make fewer mistakes. But training has no value if we can't retrieve the information we're taught. Training reinforcement is essential to ensure that knowledge and skills learned in training are applied on the job. If you are new to training reinforcement or a bit unfamiliar, here are seven key things to know.

Accenture has announced the launch of new services for testing artificial intelligence (AI) systems, powered by a "Teach and Test" methodology designed to help companies build, monitor and measure reliable AI systems within their own infrastructure or in the cloud. Accenture's "Teach and Test" methodology ensures that AI systems are producing the right decisions in two phases. The "Teach" phase focuses on the choice of data, models and algorithms that are used to train machine learning. This phase experiments and statistically evaluates different models to select the best performing model to be deployed into production, while avoiding gender, ethnic and other biases, as well as ethical and compliance risks. During the "Test" phase, AI system outputs are compared to key performance indicators, and assessed for whether the system can explain how a decision or outcome was determined.

Teaching a robot how to walk and jump is one thing, but how about a bot with a knack for dragging its backside across the floor or sliding forward on its nose. These rather unusual behaviors emerged during Google's testing of a new technique to allow simulated robots to independently learn useful skills, such as walking and jumping, which can be applied to a wide variety of tasks. "Conveniently, because skills are learned without a priori knowledge of the task, the learned skills can be used for many different tasks," the researchers note in the paper. SEE: IT leader's guide to the blockchain (Tech Pro Research) The researchers from the Google Brain team and UC Berkeley say their new approach to unsupervised learning can lead to novel and amusing behaviors, such as the simulated cheetah scooting forward on its nose, sliding forward on its backside or running backwards. "This observation motivates our claim that human designers simply do not know or cannot imagine some skills agents can acquire," the paper states.

An economic boom just dropped on the world -- and most, no doubt, aren't even aware. China's retail and technology conglomerate, Alibaba, developed an artificial intelligence model that beat the humans it competed against in a Stanford University reading and comprehension test. "This is the first time that a machine has outperformed humans on such a test," Alibaba said in a statement. Microsoft Research Asia, shortly after Alibaba's breakthrough, announced its own AI had beaten the humans on the same reading and comp test, as well. The news generated a small media earthquake through the fields of technology and AI.