We introduce a framework based on bilevel programming that unifies gradient-based hyperparameter optimization and meta-learning. We show that an approximate version of the bilevel problem can be solved by taking into explicit account the optimization dynamics for the inner objective. Depending on the specific setting, the outer variables take either the meaning of hyperparameters in a supervised learning problem or parameters of a meta-learner. We provide sufficient conditions under which solutions of the approximate problem converge to those of the exact problem. We instantiate our approach for meta-learning in the case of deep learning where representation layers are treated as hyperparameters shared across a set of training episodes. In experiments, we confirm our theoretical findings, present encouraging results for few-shot learning and contrast the bilevel approach against classical approaches for learning-to-learn.

We prove that the evolution of weight vectors in online gradient descent can encode arbitrary polynomial-space computations, even in the special case of soft-margin support vector machines. Our results imply that, under weak complexity-theoretic assumptions, it is impossible to reason efficiently about the fine-grained behavior of online gradient descent.

In compressed sensing, a small number of linear measurements can be used to reconstruct an unknown signal. Existing approaches leverage assumptions on the structure of these signals, such as sparsity or the availability of a generative model. A domain-specific generative model can provide a stronger prior and thus allow for recovery with far fewer measurements. However, unlike sparsity-based approaches, existing methods based on generative models guarantee exact recovery only over their support, which is typically only a small subset of the space on which the signals are defined. We propose Sparse-Gen, a framework that allows for sparse deviations from the support set, thereby achieving the best of both worlds by using a domain specific prior and allowing reconstruction over the full space of signals. Theoretically, our framework provides a new class of signals that can be acquired using compressed sensing, reducing classic sparse vector recovery to a special case and avoiding the restrictive support due to a generative model prior. Empirically, we observe consistent improvements in reconstruction accuracy over competing approaches, especially in the more practical setting of transfer compressed sensing where a generative model for a data-rich, source domain aids sensing on a data-scarce, target domain.

Researchers have shown for the first time that a form of artificial intelligence or machine learning known as a deep learning convolutional neural network (CNN) is better than experienced dermatologists at detecting skin cancer. In a study published in the leading cancer journal Annals of Oncology today (Tuesday), researchers in Germany, the USA and France trained a CNN to identify skin cancer by showing it more than 100,000 images of malignant melanomas (the most lethal form of skin cancer), as well as benign moles (or nevi). They compared its performance with that of 58 international dermatologists and found that the CNN missed fewer melanomas and misdiagnosed benign moles less often as malignant than the group of dermatologists. A CNN is an artificial neural network inspired by the biological processes at work when nerve cells (neurons) in the brain are connected to each other and respond to what the eye sees. The CNN is capable of learning fast from images that it "sees" and teaching itself from what it has learned to improve its performance (a process known as machine learning).

Decisions about war have always been made by humans, but now intelligent machines are on the cusp of changing things - with dramatic consequences for international affairs. This book explores the evolutionary origins of human strategy, and makes a provocative argument that Artificial Intelligence will radically transform the nature of war by changing the psychological basis of decision-making about violence. Strategy, Evolution, and War is a cautionary preview of how Artificial Intelligence (AI) will revolutionize strategy more than any development in the last three thousand years of military history. Kenneth Payne describes strategy as an evolved package of conscious and unconscious behaviors with roots in our primate ancestry. Our minds were shaped by the need to think about warfare--a constant threat for early humans.

The odds of responding well to "intensifying" antidiabetic regimens with an additional antihyperglycemic and of avoiding episodes of severe hypoglycemia could be increased by promising approaches in big data, machine learning, and real-time informatics, according to recent presentations at the American Diabetes Association (ADA) 78th Scientific Sessions, Orlando, Florida. The decision to add a glucagon-like peptide-1 receptor agonist (GLP-1 RA) to basal insulin and other oral antihyperglycemic agents that have failed to adequately control a patient's type 2 diabetes (T2DM) could be better informed, for example, with analysis of a range of patient characteristics including the other medications and dosages, and the severity and duration of diabetic symptoms and of concurrent conditions. Big-data algorithms might be used to consider these multiple parameters, and to possibly identify optimal patient characteristics for the new drug therapy, according to Esther Zimmermann, PhD, Novo Nordisk, Søborg, Denmark. "Machine learning is a new tool used for the analysis of big data that has the potential to identify trends and predict outcomes," Zimmermann explained, in describing her study. "The aim of this study was to use machine learning for extensive analysis of big, complex to data to, one, characterize patients on basal insulin for whom a GLP-1 RA was additionally prescribed and, two, identify predictors of 1 percent (or greater) reduction in A1c in (those) patients."

Instagram shows bizarre'all caught up' notification to its most prolific users Instagram is sending out a strange "all caught up" notification to its most prolific users. Anyone who scrolls all the way through their posts – so that they have seen every single one posted over the last two days – will see the message. "You're all caught up," the message will read. Underneath that will be the posts users have already seen as well as those from older than two days ago. The I.F.O. is fuelled by eight electric engines, which is able to push the flying object to an estimated top speed of about 120mph.

The flagman waves the chequered flag as Red Bull's Dutch driver Max Verstappen crosses the finish of the Austrian Formula One Grand Prix in Spielberg, central Austria, on July 1, 2018. Formula 1 plans to use cloud technology and machine learning to deliver more engaging statistic and even predictions to fans watching races on television and on its digital platforms. Cloud giant Amazon Web Services (AWS) has been signed up as an official technology partner, with its technology used to crunch the data and deliver it in a more meaningful way to fans and commentators. Each Formula 1 car produces huge amounts of data that the teams use to optimise their strategies and it is this database that Liberty Media believes can be turned into something valuable for the audience. After all, this is a sport that claims to have been'doing' big data since before the term was coined. Data scientists are using 65 years' worth of historical race data to train deep learning models that can make race predictions and give fans an insight into why a team has adopted a particular strategy.

Kajita, who studied theoretical physics at university, lived in Italy when her husband, also a researcher, was transferred to the country, and often fell prey to local pickpockets. She realized that without the knowledge of local situations, people are vulnerable to crimes. She then came up with the idea of applying her research method of explaining natural phenomena in mathematical formula to crime-prevention efforts. Kajita paid attention to email information provided by the Metropolitan Police Department on molestation cases, thefts and suspicious people. Despite the limited amount of data, predictions mostly matched areas where crimes were actually committed.

With Wimbledon underway and tennis fever threatening to shove aside World Cup fever as the foremost sporting-based malady, what better way to mark the occasion with a new iteration of Mario Tennis? The Switch version of Nintendo's popular sports game sees Mario and his friends, enemies and miscellaneous together again for more inexplicable tennis action. Players can choose from the good (Mario, Luigi, Toad, Peach), the bad (Bowser, Boo, Wario) and the just plain misunderstood (Chain Chomp). You even have the rare opportunity to play as Waluigi. As with most Nintendo titles, a big question is one you'll be faced with early on: is it any good as a single player game?