-- Model free reinforcement learning suffers from the high sampling complexity inherent to robotic manipulation or locomotion tasks. Most successful approaches typically use random sampling strategies which leads to slow policy convergence. In this paper we present a novel approach for efficient exploration that leverages previously learned tasks. We exploit the fact that the same system is used across many tasks and build a generative model for exploration based on data from previously solved tasks to improve learning new tasks. The approach also enables continuous learning of improved exploration strategies as novel tasks are learned. Extensive simulations on a robot manipulator performing a variety of motion and contact interaction tasks demonstrate the capabilities of the approach. In particular, our experiments suggest that the exploration strategy can more than double learning speed, especially when rewards are sparse. Moreover, the algorithm is robust to task variations and parameter tuning, making it beneficial for complex robotic problems. I. INTRODUCTION Deep reinforcement learning has attracted a lot of attention for robotic applications where full robot models can be difficult to identify, especially for contact dynamics, and lead to computationally challenging planning and control problems.

Abstract: Sepsis, a dysregulated immune system response to infection, is among the leading causes of morbidity, mortality, and cost overruns in the Intensive Care Unit (ICU). Ear ly prediction of sepsis can improve situational awareness amongst clinicians and facilitate timely, protective interventions. While the application of predictive analytics in ICU patients has shown early promising results, much of the work has been encumbe red by high false - alarm rates. Efforts to improve specificity have been limited by several factors, most notably the difficulty of labeling sepsis onset time and the low prevalence of septic - events in the ICU. We show that by coupling a clinical criterion for defining sepsis onset time with a treatment policy (e.g., initiation of antibiotics within one hour of meeting the criterion), one may rank the relative utility of various criteria through offline policy evaluation. Given the optimal criterion, DeepAISE automatically learns predictive features related to higher - order interactions and temporal patterns among clinic al risk factors that maximize the data likelihood of observed time to septic events. DeepAISE has been incorporated into a clinical workflow, which provides real - time hourly sepsis risk scores. A comparative study of four baseline models indicates that Dee pAISE produces the most accurate predictions (AUC 0.90 and 0.87) and the lowest false alarm rates (FAR 0.20 and 0.26) in two separate cohorts (internal and external, respectively), while simultaneously producing interpretable representations of the clinica l time series and risk factors. Introduction Sepsis is a syndromic, life - threatening condition that arises when the body's response to infection injures its own internal organs (1) . Though the condition lacks the same public notoriety as other conditions like heart attacks, 6% of all hospitalized patients in the U nited S tates carry a primary diagnosis of sepsis as compared to 2.5% for the latter (2) . When all hospital deaths are ultimately considered, nearly 35% are attributable to sepsis (2) . This condition stands in stark contrast to heart attacks which have a mortality rate of 2.7 - 9.6% and only cost the US $12.1 billion ann ually, roughly half of the cost of sepsis (3) .

We study the problem of rare event prediction for a class of slow-fast nonlinear dynamical systems. The state of the system of interest is described by a slow process, whereas a faster process drives its evolution. By taking advantage of recent advances in machine learning, we present a data-driven method to predict the future evolution of the state. We show that our method is capable of predicting a rare event at least several time steps in advance. We demonstrate our method using numerical experiments on two examples and discuss the mathematical and broader implications of our results.

Reinforcement learning algorithms, though successful, tend to over-fit to training environments hampering their application to the real-world. This paper proposes WR$^{2}$L; a robust reinforcement learning algorithm with significant robust performance on low and high-dimensional control tasks. Our method formalises robust reinforcement learning as a novel min-max game with a Wasserstein constraint for a correct and convergent solver. Apart from the formulation, we also propose an efficient and scalable solver following a novel zero-order optimisation method that we believe can be useful to numerical optimisation in general. We contribute both theoretically and empirically. On the theory side, we prove that WR$^{2}$L converges to a stationary point in the general setting of continuous state and action spaces. Empirically, we demonstrate significant gains compared to standard and robust state-of-the-art algorithms on high-dimensional MuJuCo environments.

From using a simple web cam to identify objects to training a network in the cloud, these resources will help you take advantage of all MATLAB has to offer for deep learning.

This article is part of Demystifying AI, a series of posts that (try to) disambiguate the jargon and myths surrounding AI. One of the most influential technologies of the past decade is artificial neural networks, the fundamental piece of deep learning algorithms, the bleeding edge of artificial intelligence. You can thank neural networks for many of applications you use every day, such as Google's translation service, Apple's Face ID iPhone lock and Amazon's Alexa AI-powered assistant. Neural networks are also behind some of the important artificial intelligence breakthroughs in other fields, such as diagnosing skin and breast cancer, and giving eyes to self-driving cars. The concept and science behind artificial neural networks have existed for many decades.

DeepMind, the U.K.-based AI research subsidiary acquired by Alphabet in 2014 for $500 million, today detailed ecological research its science team is conducting to develop AI systems that'll help study the behavior of animal species in Tanzania's Serengeti National Park. It hopes to expedite the analysis of data from hundreds of motion-detecting field cameras, which have captured millions of images since they were deployed by the Serengeti Lion Research program over nine years ago. "The Serengeti is one of the last remaining sites in the world that hosts an intact community of large mammals … As human encroachment around the park becomes more intense, these species are forced to alter their behaviours in order to survive," wrote DeepMind in a blog post. "Increasing agriculture, poaching, and climate abnormalities contribute to changes in animal behaviors and population dynamics, but these changes have occurred at spatial and temporal scales which are difficult to monitor using traditional research methods." For nearly a decade, conservationists have tapped the aforementioned cameras to keep tabs on animals within the park's core, enabling them to study their distribution and demography.

Deep learning techniques do a good job at building models by correlating data points. But many AI researchers believe that more work needs to be done to understand causation and not just correlation. The field of causal deep learning -- useful in determining why something happened -- is still in its infancy, and it is much more difficult to automate than neural networks. Much of AI is about finding hidden patterns in large amounts of data. Soumendra Mohanty, executive vice president and chief data analytics officer at L&T Infotech, a global IT service company, said, "Obviously, this aspect drives us to the'what,' but rarely do we go down the path of understanding the'why.'"

Complex algorithms utilized in data analytics are called unsupervised machine learning, but image recognition or besting the Go champion utilizes supervised machine learning; technology that utilizes neural networks. "The new platform is expected to test algorithms that use an advanced form of AI called deep learning, a technique that has the potential to identify more complex patterns than traditional machine-learning algorithms. "It's a massive breakthrough for us," Mr. Taneja said. Visa currently uses machine-learning algorithms to sift through data to identify anomalies, an effort that prevents billions of dollars in fraudulent transactions annually, Mr. Taneja said. One such Visa fraud-detection system, Advanced Authorization, prevented about $25 billion in fraud in the year ended April 30, according to the company. But the current models have limitations. Researchers must know the signals that might indicate fraud--such as a purchase taking place at an unusual time of day--and write the rules to tell the model what to do when it identifies suspicious activity. Criminal activity sometimes slips by unnoticed because hackers are getting more sophisticated at evading the warning signs that current machine-learning models are trying to detect. Deep-learning models can automatically identify more complex patterns by themselves. For example, if a customer uses his or her card in another country for the first time, deep-learning algorithms will be able to tell, with more accuracy and fewer false positives than traditional machine learning, whether it's a legitimate transaction. The algorithms will be able to take into account previous transactions at airlines and hotels, as long as they are made with Visa cards."

Deep Learning with R introduces the world of deep learning using the powerful Keras library and its R language interface. The book builds your understanding of deep learning through intuitive explanations and practical examples.