Stochastic gradient descent (SGD) (Robbins and Monro, 1951) and its variants have been widely used in training deep neural networks. Among those variants, adaptive gradient methods (AdaGrad) (Duchi et al., 2011; McMahan and Streeter, 2010), which scale each coordinate of the gradient by a function of past gradients, can achieve better performance than vanilla SGD in practice when the gradients are sparse. An intuitive explanation for the success of AdaGrad is that it automatically adjusts the learning rate for each feature based on the partial gradient, which accelerates the convergence. However, AdaGrad was later found to demonstrate degraded performance especially in cases where the loss function is nonconvex or the gradient is dense, due to rapid decay of learning rate.

In this paper we address the memory demands that come with the processing of 3-dimensional, high-resolution, multi-channeled medical images in deep learning. We exploit memory-efficient backpropagation techniques, to reduce the memory complexity of network training from being linear in the network's depth, to being roughly constant $ - $ permitting us to elongate deep architectures with negligible memory increase. We evaluate our methodology in the paradigm of Image Quality Transfer, whilst noting its potential application to various tasks that use deep learning. We study the impact of depth on accuracy and show that deeper models have more predictive power, which may exploit larger training sets. We obtain substantially better results than the previous state-of-the-art model with a slight memory increase, reducing the root-mean-squared-error by $ 13\% $. Our code is publicly available.

Short-term load forecasting is one of the crucial sections in smart grid. Precise forecasting enables system operators to make reliable unit commitment and power dispatching decisions. With the advent of big data, a number of artificial intelligence techniques such as back propagation, support vector machine have been used to predict the load of the next day. Nevertheless, due to the noise of raw data and the randomness of power load, forecasting errors of existing approaches are relatively large. In this study, a short-term load forecasting method is proposed on the basis of empirical mode decomposition and long short-term memory networks, the parameters of which are optimized by a particle swarm optimization algorithm. Essentially, empirical mode decomposition can decompose the original time series of historical data into relatively stationary components and long short-term memory network is able to emphasize as well as model the timing of data, the joint use of which is expected to e ffectively apply the characteristics of data itself, so as to improve the predictive accuracy. The e ffectiveness of this research is exemplified on a realistic data set, the experimental results of which show that the proposed method has higher forecasting accuracy and applicability, as compared with existing methods. Introduction Based on historical data, power load forecasting is to explore the developing law of electricity, establish models between power demand and features, then make a valid prediction of future load [1]. A lot of operations in power systems sharply depend on the future information provided by predictions, for example making a satisfying unit commitment (UC) decision [2], saving energy and reducing the cost of power generation [3].

Children are far more susceptible than adults to being influenced by robots, according to a study. Researchers at the University of Plymouth used a technique developed in the 1950s to determine how much influence robots can have on people's opinions. The Asch paradigm was originally used to describe how people will usually follow the opinions of others, even if they are clearly wrong. "People often follow the opinions of others and we've known for a long time that it is hard to resist taking over views and opinions of people around us," said robotics professor Tony Belpaeme, who led the study alongside Plymouth researcher Anna Vollmer. "We know this as conformity. But as robots will soon be found in the home and the workplace, we were wondering if people would conform to robots. "What our results show is that adults do not conform to what the robots are saying.

LG Electronics India launched the much-anticipated range of televisions in India featuring Artificial Intelligence (AI) ThinQ. Designed to bring a new level of convenience, enhanced connectivity and a more immersive TV viewing experiences, the new range includes various models under its OLED, Super UHD, UHD and Smart TV category. With AI functionality in LG TVs, the consumers can directly speak into the remote to control TV functions and seamlessly discover and play content. These TVs doesn't only work on fixed voice commands but also understand the intent of the query before providing a search result. The TV not only Listens and Answers but Listens, thinks and Answer.

Reinventing the future of work can lead to shared prosperity. An artificial intelligence-driven career adviser, an industrial smart glove, freelance insurance, a tactile laptop for the visually impaired. The 12 European finalists of the global MIT Inclusive Innovation Challenge are "improving economic opportunity for workers," according to the MIT Initiative on the Digital Economy. The challenge is the flagship program of the initiative, and this year the initiative launched a worldwide competition divided into five regions: North America, Latin America, Europe, Africa, and Asia. "If we employ inclusive innovation globally, it could be the best thing that ever happened to humanity," Erik Brynjolfsson, director of the initiative, said in a statement.

Exponential advancements in technology within the last half century have profoundly reshaped humanity and continue to do so continuously. Concepts which once seemed as fantasy Sci-Fi, visualized through Hollywood hits such as The Terminator (1984) and Eagle Eye (2008) have steadily and inconspicuously become a part of our reality. More recently, the futurist show, Black Mirror (2011) featured on Netflix gives us a glimpse of what the future may hold. One thing in common for all of these shows is the portrayal of possibilities with regards to advancements in computer technology -- be it in the form of a highly intelligent, autonomous, sophisticated robot like the Terminator (with a massive capacity for destruction)or ARIIA, a supercomputer able to manipulate almost all connected devices and command its victims to fulfill its agenda. Artificial Intelligence (AI) is seen to be the core driver of current trends within the tech sector and has vastly developed since the term was first coined in the 1950's. It is embedded in our phones, in the form of online chat bots and as phone operators to name a few contemporary use cases.

The stern of a US destroyer that was blown off the ship by a Japanese mine 75 years ago, killing 71, has been found off Alaska. The fragment of the USS Abner Read was found in the Bering Sea off the Aleutian island of Kiska, where it sank after being torn off by an explosion while conducting an anti-submarine patrol. The remaining crew managed to save the ship, which was repaired after the attack. On July 17, a NOAA-funded team of scientists from Scripps Institution of Oceanography at the University of California San Diego and the University of Delaware discovered the missing 75- foot stern section in 290 feet of water off of Kiska, one of only two United States territories to be occupied by foreign forces in the last 200 years. After sonar mounted to the side of the research ship Norseman II identified a promising target, the team sent down a deep-diving, remotely operated vehicle to capture live video for confirmation.

The annual conference is a showcase of China's burgeoning robot industry ranging from companion robots to those deployed on manufacturing assembly line and entertainment.

If 2017 was the year artificial intelligence rose to prominence, 2018 is when we're seeing it go mainstream. Whichever area you work in, it's likely AI will become increasingly prevalent in your everyday activity. Wherever you are in the world – whether you are an expert in AI, someone whose job increasingly uses AI or simply an interested reader we would like to hear from you. Earlier this year, the Guardian published a long read that asked: Has technology evolved beyond our control? Its author, James Bridle, argued that "our technologies are extensions of ourselves, codified in machines and infrastructures, in frameworks of knowledge and action. Computers are not here to give us all the answers, but to allow us to put new questions, in new ways, to the universe."