Many attempts have been done to extend the great success of convolutional neural networks (CNNs) achieved on high-end GPU servers to portable devices such as smart phones. Providing compression and acceleration service of deep learning models on the cloud is therefore of significance and is attractive for end users. However, existing network compression and acceleration approaches usually fine-tuning the svelte model by requesting the entire original training data (\eg ImageNet), which could be more cumbersome than the network itself and cannot be easily uploaded to the cloud. In this paper, we present a novel positive-unlabeled (PU) setting for addressing this problem. In practice, only a small portion of the original training set is required as positive examples and more useful training examples can be obtained from the massive unlabeled data on the cloud through a PU classifier with an attention based multi-scale feature extractor. We further introduce a robust knowledge distillation (RKD) scheme to deal with the class imbalance problem of these newly augmented training examples. The superiority of the proposed method is verified through experiments conducted on the benchmark models and datasets. We can use only $8\%$ of uniformly selected data from the ImageNet to obtain an efficient model with comparable performance to the baseline ResNet-34.

Sepsis is the leading cause of death in non-coronary intensive care units. Moreover, a delay of antibiotic treatment of patients with severe sepsis by only few hours is associated with increased mortality. This insight makes accurate models for early prediction of sepsis a key task in machine learning for healthcare. Previous approaches have achieved high AUROC by learning from electronic health records where sepsis labels were defined automatically following established clinical criteria. We argue that the practice of incorporating the clinical criteria that are used to automatically define ground truth sepsis labels as features of severity scoring models is inherently circular and compromises the validity of the proposed approaches. We propose to create an independent ground truth for sepsis research by exploiting implicit knowledge of clinical practitioners via an electronic questionnaire which records attending physicians' daily judgements of patients' sepsis status. We show that despite its small size, our dataset allows to achieve state-of-the-art AUROC scores. An inspection of learned weights for standardized features of the linear model lets us infer potentially surprising feature contributions and allows to interpret seemingly counterintuitive findings.

--The purpose of this work is to develop and study a distributed strategy for Pareto optimization of an aggregate cost consisting of regularized risks. Each risk is modeled as the expectation of some loss function with unknown probability distribution while the regularizers are assumed deterministic, but are not required to be differentiable or even continuous. The individual, regularized, cost functions are distributed across a strongly-connected network of agents and the Pareto optimal solution is sought by appealing to a multi-agent diffusion strategy. T o this end, the regularizers are smoothed by means of infimal convolution and it is shown that the Pareto solution of the approximate, smooth problem can be made arbitrarily close to the solution of the original, non-smooth problem. Performance bounds are established under conditions that are weaker than assumed before in the literature, and hence applicable to a broader class of adaptation and learning problems. Index T erms --Distributed optimization, diffusion strategy, smoothing, proximal operator, non-smooth regularizer, proximal diffusion, regularized diffusion. The objective of distributed learning is the solution of global, stochastic optimization problems across networks of agents through localized interactions and without information about the statistical properties of the data. Using streaming data, the resulting strategies are adaptive in nature and able to track drifts in the location of the minimizers due to variations in the statistical properties of the data. Regularization is one useful technique to encourage or enforce structural properties on the sought after minimizer, such as sparsity or constraints. A substantial number of regularizers are inherently non-smooth, while many cost functions are differentiable.

Differentiable Architecture Search (DARTS) has attracted a lot of attention due to its simplicity and small search costs achieved by a continuous relaxation and an approximation of the resulting bi-level optimization problem. However, DARTS does not work robustly for new problems: we identify a wide range of search spaces for which DARTS yields degenerate architectures with very poor test performance. We study this failure mode and show that, while DARTS successfully minimizes validation loss, the found solutions generalize poorly when they coincide with high validation loss curvature in the space of architectures. We show that by adding one of various types of regularization we can robustify DARTS to find solutions with smaller Hessian spectrum and with better generalization properties. Based on these observations we propose several simple variations of DARTS that perform substantially more robustly in practice. Our observations are robust across five search spaces on three image classification tasks and also hold for the very different domains of disparity estimation (a dense regression task) and language modelling. We provide our implementation and scripts to facilitate reproducibility.

Driving in dynamically changing traffic is a highly challenging task for autonomous vehicles, especially in crowded urban roadways. The Artificial Intelligence (AI) system of a driverless car must be able to arbitrate between different driving strategies in order to properly plan the car's path, based on an understandable traffic scene model. In this paper, an AI behavior arbitration algorithm for Autonomous Driving (AD) is proposed. The method, coined AIBA (AI Behavior Arbitration), has been developed in two stages: (i) human driving scene description and understanding and (ii) formal modelling. The description of the scene is achieved by mimicking a human cognition model, while the modelling part is based on a formal representation which approximates the human driver understanding process. The advantage of the formal representation is that the functional safety of the system can be analytically inferred. The performance of the algorithm has been evaluated in Virtual Test Drive (VTD), a comprehensive traffic simulator, and in GridSim, a vehicle kinematics engine for prototypes.

Walgreens is getting its wings. The pharmacy chain has teamed up with Alphabet's drone delivery service Wing to bring food and beverage, over-the-counter medication and other items to consumers. This'store to door' testing is set to begin next month in Virginia and will offer more than 100 products and pre-built'packs' for purchase in the Wing app. Walgreens has teamed up with Alphabet's drone delivery service Wing to bring food and beverage, over-the-counter medication and other items to consumers The partnership between Walgreens and Wing aims to further explore the future of health and wellness products and retail delivery through the air, offering product availability and home delivery minutes after placing orders via the Wing app. 'Walgreens continues to explore partnerships to transform and modernize our customer experience and we are proud to be the first retailer in the U.S. to offer an on-demand commercial drone delivery option with Wing,' said Vish Sankaran, chief innovation officer, Walgreens Boots Alliance, Inc., in a press release.

Speaking at his mid-term semiconductor industry forecast seminar in London this week, Malcolm Penn, chairman and CEO of industry analyst Future Horizons, assured attendees that industry fundamentals were sound, and after a fall of around 15% in 2019, the industry will rebound with around 4% revenue growth to $414 billion in 2020. He said, "The fundamentals are sound. In terms of IC unit growth, fab capacity and average selling price, they are all in in good shape. It's the timing of the upswing in the economy that puts it into doubt." He added, "Rebound is a certainty, but its timing is not."

A Class 6 student, Samaira Mehta already seems to be on top of her game. Of Indian origin and based in California, this 11-year-old girl in tech is an inventor, and has invented CoderBunnyz, a STEM coding board game to teach coding to kids between the age group of four and 10. Samaira has taken Silicon Valley by storm and has been a part of more than 50 conferences. She has held 60 workshops that spotlight her board game, and taught over 2,000 kids, including over 50 "Google kids" at Googleplex, Google's headquarters in Mountain View. The young girl also received a letter from the White House, from then First Lady Michelle Obama, for her work.

The dreaded El Niño strikes the globe every 2 to 7 years. As warm waters in the tropical Pacific Ocean shift eastward and trade winds weaken, the weather pattern ripples through the atmosphere, causing drought in southern Africa, wildfires in South America, and flooding on North America's Pacific coast. Climate scientists have struggled to predict El Niño events more than 1 year in advance, but artificial intelligence (AI) can now extend forecasts to 18 months, according to a new study. The work could help people in threatened regions better prepare for droughts and floods, for example by choosing which crops to plant, says William Hsieh, a retired climate scientist in Victoria, Canada, who worked on early El Niño forecasts but who was not involved in the current study. Longer forecasts could have "large economic benefits," he says.

Wipro signed a strategic partnership with the Industrie 4.0 Maturity Center (I4.0MC), Based in Aachen, Germany, I4.0MC is a part of RWTH Aachen Campus With the help of this strategic partnership, Wipro consultants use the i4.0MC's program to support their client's digital transformation processes, the announcement notes. The acatech Industrie 4.0 Maturity Index, applied by the Industrie 4.0 Maturity Center, works as a methodical guideline to individually design the path to an agile company and to derive the necessary steps. The partnership will also promote the collaboration between the industry and academic experts across the industries such as industrial manufacturing, consumer goods, automotive, oil & gas, and life science. Christian Hocken, MBA, Managing Partner said, "We are looking forward to the cooperation with a leading technology company. Our competences complement each other in an ideal way. We are providing the management frameworks and tools while Wipro will be realizing the digital transformation. Together we will be able to serve our customers with tailor-made transformation projects to become a data-driven, agile company."