Current newspapers are full of horrific tales of "cyber-attackers" threatening our energy systems. And, if not for the notorious "evil state"-actor, it is the ongoing digitization necessary to enable increasing renewable and volatile energy generation that threatens our energy supply and thus the stability of our society. And while the main approach seems to be to patch-up the detected vulnerabilities of protocols, software and controller devices, our approach is to research and develop the means to systematically design and test systems that are structurally resilient against failures and attackers alike. Security in cyber-systems mostly should be concerned with establishing asymetric control in favour of the operator of a system. In order to achieve this on a structural level at design time, reproducible benchmark tests are required.

Another tech company doing something it said it wouldn't. Another eye roll, another shrug? On Tuesday, the London-based artificial intelligence company DeepMind announced that the team behind Streams – an app designed to monitor people in hospital with kidney disease – will be joining DeepMind's sister company Google. The tech giant wants to turn Streams into an AI-powered assistant for doctors and nurses. To create Streams, DeepMind used identifiable medical records of 1.6 million people obtained in a deal with the Royal …

Privacy campaigners have raised fears for patient data following Google's takeover of an artificial intelligence health app used in NHS hospitals. London-based AI firm DeepMind said its Streams app will be subsumed by the technology giant in a move that one expert described as "totally unacceptable" and a betrayal to patient's privacy. DeepMind, which is owned by Google but has operated the app independently until now, justified the decision in a blog post that explained how Google would allow the app to scale in a way that would not be possible by itself. The app uses AI to provide doctors and nurses with an easy-access dashboard of patients' medical records. "Our vision is for Streams to now become an AI-powered assistant for nurses and doctors everywhere – combining the best algorithms with intuitive design, all backed up by rigorous evidence," the post stated.

Google looks to be getting a firmer hold on NHS patient data by absorbing its DeepMind Health AI lab - a leading UK health technology developer. The news has raised concerns about the privacy of NHS patient's data which is used by DeepMind and could now be commercialised by Google. DeepMind was bought by Google's parent company Alphabet for £400 million ($520m) in 2014 and up until now has maintained independence. Now the London-based lab will be sharing operations with the US-based Google Health unit. It was created after Google bought University College London spinout, DeepMind, for £400 million in 2014.

Google has been accused of breaking promises to patients, after the company announced it would be moving a healthcare-focused subsidiary, DeepMind Health, into the main arm of the organisation. The restructure, critics argue, breaks a pledge DeepMind made when it started working with the NHS that "data will never be connected to Google accounts or services". The change has also resulted in the dismantling of an independent review board, created to oversee the company's work with the healthcare sector, with Google arguing that the board was too focused on Britain to provide effective oversight for a newly global body. Google says the restructure is necessary to allow DeepMind's flagship health app, Streams, to scale up globally. The app, which was created to help doctors and nurses monitor patients for AKI, a severe form of kidney injury, has since grown to offer a full digital dashboard for patient records.

Semiconductor Engineering sat down to discuss artificial intelligence (AI), machine learning, and chip and photomask manufacturing technologies with Aki Fujimura, chief executive of D2S; Jerry Chen, business and ecosystem development manager at Nvidia; Noriaki Nakayamada, senior technologist at NuFlare; and Mikael Wahlsten, director and product area manager at Mycronic. What follows are excerpts of that conversation. To read part one, click here. SE: Artificial neural networks, the precursor of machine learning, was a hot topic in the 1980s. In neural networks, a system crunches data and identifies patterns.

Alphabet is shuffling some of its companies around as it works to better organize the health projects that are currently spread across its subsidiaries. So going forward, DeepMind's health unit will instead exist under the Google umbrella and it will be part of the company's recently formed Google Health initiative. Specifically, DeepMind's Streams app, which physicians in the UK have used to help treat their patients, will be moving over to Google, and the Google Health team will be working on expanding the app to more regions. We're excited to announce that the team behind Streams - our app supporting doctors and nurses to deliver faster, better care to patients - will be joining Google. Google recently brought in David Feinberg to lead the new Google Health group, with the goal of organizing Alphabet's health efforts and enhancing collaborations across its subsidiaries.

Generating musical audio directly with neural networks is notoriously difficult because it requires coherently modeling structure at many different timescales. Fortunately, most music is also highly structured and can be represented as discrete note events played on musical instruments. Herein, we show that by using notes as an intermediate representation, we can train a suite of models capable of transcribing, composing, and synthesizing audio waveforms with coherent musical structure on timescales spanning six orders of magnitude ( 0.1 ms to 100 s), a process we call Wave2Midi2Wave. This large advance in the state of the art is enabled by our release of the new MAESTRO (MIDI and Audio Edited for Synchronous TRacks and Organization) dataset, composed of over 172 hours of virtuosic piano performances captured with fine alignment ( 3 ms) between note labels and audio waveforms. The networks and the dataset together present a promising approach toward creating new expressive and interpretable neural models of music. Since the beginning of the recent wave of deep learning research, there have been many attempts to create generative models of expressive musical audio de novo. These models would ideally generate audio that is both musically and sonically realistic to the point of being indistinguishable to a listener from music composed and performed by humans. However, modeling music has proven extremely difficult due to dependencies across the wide range of timescales that give rise to the characteristics of pitch and timbre (short-term) as well as those of rhythm (medium-term) and song structure (long-term). On the other hand, much of music has a large hierarchy of discrete structure embedded in its generative process: a composer creates songs, sections, and notes, and a performer realizes those notes with discrete events on their instrument, creating sound.

Contemporary smart power grids are being challenged by rapid voltage fluctuations, due to large-scale deployment of renewable generation, electric vehicles, and demand response programs. In this context, monitoring the grid's operating conditions in real time becomes increasingly critical. With the emergent large scale and nonconvexity however, past optimization based power system state estimation (PSSE) schemes are computationally expensive or yield suboptimal performance. To bypass these hurdles, this paper advocates deep neural networks (DNNs) for real-time power system monitoring. By unrolling a state-of-the-art prox-linear SE solver, a novel modelspecific DNN is developed for real-time PSSE, which entails a minimal tuning effort, and is easy to train. To further enable system awareness even ahead of the time horizon, as well as to endow the DNN-based estimator with resilience, deep recurrent neural networks (RNNs) are pursued for power system state forecasting. Deep RNNs exploit the long-term nonlinear dependencies present in the historical voltage time series to enable forecasting, and they are easy to implement. Numerical tests showcase improved performance of the proposed DNN-based estimation and forecasting approaches compared with existing alternatives. Empirically, the novel model-specific DNN-based PSSE offers nearly an order of magnitude improvement in performance over competing alternatives, including the widely adopted Gauss-Newton PSSE solver, in our tests using real load data on the IEEE 118-bus benchmark system.

Automated melodic phrase detection and segmentation is a classical task in content-based music information retrieval and also the key towards automated music structure analysis. However, traditional methods still cannot satisfy practical requirements. In this paper, we explore and adapt various neural network architectures to see if they can be generalized to work with the symbolic representation of music and produce satisfactory melodic phrase segmentation. The main issue of applying deep-learning methods to phrase detection is the sparse labeling problem of training sets. We proposed two tailored label engineering with corresponding training techniques for different neural networks in order to make decisions at a sequential level. Experiment results show that the CNN-CRF architecture performs the best, being able to offer finer segmentation and faster to train, while CNN, Bi-LSTM-CNN and Bi-LSTM-CRF are acceptable alternatives.