Singtel has inked a five-year pact with Nanyang Technological University (NTU) to bolster Singapore's capabilities in artificial intelligence (AI), the internet-of-things (IoT), data analytics and robotics. You forgot to provide an Email Address. This email address doesn't appear to be valid. This email address is already registered. You have exceeded the maximum character limit.

NEW DELHI: Bennett University"s computer science and engineering (CSE) department held its first international conference on machine learning and data science at its Greater Noida campus that saw researchers and academicians deliberating on the new wave of technologies and their impact on the world of big data, machine learning and artificial intelligence (AI). The two-day conference was organised by the department in association with the Institute of Electrical and Electronics Engineers (IEEE) Computer Society. Bennett University has been set up by the Times Group, which publishes ET. "We are pleased to be in India to establish a relationship with all the major universities including Bennett University, so that we can develop and nurture the relationship such that India becomes a very important player in the computer society," said Roger Fujii, 2016 president of the society. The event saw attendance from 50 organisations including IBM, Nvidia, Tata Consultancy Services, Wipro, Accenture, Dell and Infosys, among others. Also attending were about 400 participants representing premier institutions such as the Indian Institutes of Technology, National Institutes of Technology, Indian Institutes of Management, Jawaharlal Nehru University, Delhi Technological University, University of Delhi, Kalinga Institute of Industrial Technology and Malaviya National Institute of Technology, Jaipur.

The scope for facial detection to be used for large-scale surveillance is only just beginning to be realised. In September this year, Moscow hooked up its CCTV network to a facial-recognition system. New York is planning to roll out facial detection across bridges and tunnels. London's facial-recognition database has been criticised for going "far beyond custody purposes", and China is taking all of this to a whole new level of total state surveillance. But the invention of the ship also led to the invention of the pirate.

The news is likely to be divisive across a dance music community already locked in terminal (and, often, insufferable) arguments as to what constitutes a real DJ. You only need to Google phrases like'sync button' for evidence. The last version of the programme had already taken major steps towards allowing anyone to mix singles seamlessly, proving popular thanks to support from Spotify and the way it uses the new MacBook Pro Touch Bar, one of the biggest changes in the Apple laptop portfolio for some time. This new update can now scan tracks to find the perfect intro and outro based on what's already playing, advising on ideal fade timings while filtering and EQing to suit, and can even search your library to make recommendations as to what will sound best next. The news comes not long after we wrote about the world's first machine learning producer arriving, SKYGGE, two months after Saudi Arabia gave a'female' robot citizenship (despite the fact women in the country can't vote) and Burning Man revealed its theme for next year's festival will be Artificial Intelligence, and in the wake of Facebook pulling the plug on developing computers that can write their own code after a language was created that nobody, aside from the machines, could understand.

Samsung isn't going to let Apple hog the smart speaker limelight early next year. A new report reveals the South Korea giant is preparing to introduce its Bixby-powered speaker in the first half of 2018, so it is now expected to go head-to-head with the Cupertino giant's HomePod. On Thursday, Bloomberg learned from people familiar with Samsung's plans that the South Korea tech company is planning to launch its Bixby-powered smart speaker in the first half of the following year. While this may seem like Samsung's entry to the voice-controlled devices market is coming in late, its release would just be enough for it to rival Apple's HomePod, which is already scheduled for launch in the first quarter. Since the Bixby-powered speaker will compete with not only the HomePod but also established smart speakers like Amazon's Echo series and Google's Google Home, Samsung is setting its new product apart from its rivals by promising a device that's strongly focused on audio quality and efficient management of connected home appliances. The Bixby speaker is also said to have the advantage of synchronizing seamlessly with Samsung TVs, Galaxy phones and other Samsung products.

BENGALURU: As firms' appetite for the adoption of Artificial Intelligence (AI) grows, a report on Thursday said 68.6 per cent Indian organisations might deploy it before 2020. The Intel India commissioned report, undertaken by the International Data Corporation (IDC) that surveyed 194 Indian organisations across sectors, said 71 per cent are looking at increased process automation as a key benefit which could drive spike spends on this technology by 2020. The report also said that nearly 75 per cent of the firms surveyed anticipate benefits in business process efficiency and employee productivity with the use of AI. While 64 per cent of the respondents believe that this technology can empower them in revenue augmentation through better targeting of offers and improved sales processes, 76 per cent of the companies are or believe that they will face a shortage of skilled personnel to harness the power of AI. "This research is a small step towards comprehending this knowledge, and enabling companies, such as ours, shape strategy and move ahead in the right direction," Prakash Mallya, Managing Director, Sales and Marketing Group, Intel India, said in a statement. Intel currently powers 97 per cent of data centre servers running AI workloads worldwide and has been investing in the development of the ecosystem in India.

The Hong Kong University of Science and Technology (HKUST) signed a Memorandum of Understanding (MOU) with Megvii (Face) yesterday on establishing a joint laboratory on artificial intelligence (AI) and computer vision. The lab will be dedicated to improving people's living and advance knowledge frontiers through researches in AI and image recognition and analysis, marking a new milestone in the collaboration between HKUST and Megvii. Artificial Intelligence is an indispensable part to our future development. As a pioneer in computer vision and deep learning, Megvii, commonly known as Face, possesses world leading hardware technology and algorithm – including "Paying with Face" and "City Skynet" which were widely used in the mainland. Complementing its strength, HKUST also has an internationally recognized profile in computer vision research, in particular its work in object and environment recognition, adding on to HKUST's competency in robotics and autonomous systems as one of the University's five research focuses, the two parties are set to create more innovative applications in AI and computer vision.

"The science of AI has no borders, neither do its benefits," Fei-Fei Li, chief scientist at Google's AI business, said in a blog post Wednesday announcing the new center. But China's internet borders are fortified by the so-called Great Firewall, and most of Google's biggest products -- its search engine, YouTube and Gmail -- have been blocked by the country's vast censorship apparatus for years. Google (GOOGL) effectively left China in 2010, but the country's 730 million internet users make it too large a market to ignore. The company has made no secret of its desire to find ways to rebuild its presence there. Related: Google's man-versus-machine showdown blocked in China Its artificial intelligence unit DeepMind teamed up with Chinese authorities to hold a five-day festival in the country earlier this year.

Music highlights are valuable contents for music services. Most methods focused on low-level signal features. We propose a method for extracting highlights using high-level features from convolutional recurrent attention networks (CRAN). CRAN utilizes convolution and recurrent layers for sequential learning with an attention mechanism. The attention allows CRAN to capture significant snippets for distinguishing between genres, thus being used as a high-level feature. CRAN was evaluated on over 32,000 popular tracks in Korea for two months. Experimental results show our method outperforms three baseline methods through quantitative and qualitative evaluations. Also, we analyze the effects of attention and sequence information on performance.

Recently, deep neural networks (DNNs) have been regarded as the state-of-the-art classification methods in a wide range of applications, especially in image classification. Despite the success, the huge number of parameters blocks its deployment to situations with light computing resources. Researchers resort to the redundancy in the weights of DNNs and attempt to find how fewer parameters can be chosen while preserving the accuracy at the same time. Although several promising results have been shown along this research line, most existing methods either fail to significantly compress a well-trained deep network or require a heavy fine-tuning process for the compressed network to regain the original performance. In this paper, we propose the \textit{Block Term} networks (BT-nets) in which the commonly used fully-connected layers (FC-layers) are replaced with block term layers (BT-layers). In BT-layers, the inputs and the outputs are reshaped into two low-dimensional high-order tensors, then block-term decomposition is applied as tensor operators to connect them. We conduct extensive experiments on benchmark datasets to demonstrate that BT-layers can achieve a very large compression ratio on the number of parameters while preserving the representation power of the original FC-layers as much as possible. Specifically, we can get a higher performance while requiring fewer parameters compared with the tensor train method.