Chinese tech giant Huawei on Wednesday released Atlas 900, the world’s fastest Artificial Intelligence (AI) training cluster which would help to make technology more readily available for different fields of scientific research and business innovation.
“The future of computing is a massive market worth more than two trillion US dollars. We will keep investing with a strategy that focuses on four key areas. We will push the boundaries of architecture, invest in processors for all scenarios, keep clear business boundaries, and build an open ecosystem,” said Ken Hu, Huawei’s Deputy Chairman.
Atlas 900 combines the power of thousands of Ascend processors. It takes Atlas 900 only 59.8 seconds to train ResNet-50, the gold standard for measuring AI training performance. This is 10 seconds faster than the previous world record, the company said in a statement.
Huawei also deployed Atlas 900 on Huawei Cloud as a cluster service, making extraordinary computing power more broadly accessible to its customers across different industries.
The company has offered these services at a great discount to universities and scientific research institutes worldwide.
Huawei estimates that in the next five years, statistical computing will become the mainstream and AI computing will account for more than 80 per cent of all computing power used around the world.
In addition, the company announced its flagship Kirin 990 System on Chip (SoC) will be available in India soon.
Governments the world over are learning new tactics to quash dissent on various Social Media platforms, responding with tweets designed to distract and confuse like longer hashtags, according to a team of political scientists.
In a study of Twitter interactions during Venezuela’s 2014 protests, in which citizens voiced opposition to government leaders and called for improvements to their standard of living, the tweets of the protesters focused mainly on the protest itself, while the tweets issued by the ruling regime covered more diverse topics.
This could mean that regimes are growing more savvy in their use of social media to help suppress mass movements.
“When we started doing this study there had been a lot of optimism about the capacity of social media to produce revolutions throughout the world, like Arab Spring and the Color Revolutions in Europe,” said Kevin Munger, assistant professor of political science and social data analytics, Penn State.
“But it seems like, in hindsight, this was the result of short-term disequilibrium between the capacity of the masses to use this technology and the limited capacity of these elites to use it.”
A lot of these elites may have not been keeping up with modern communication technology and got caught unawares.
So, for that short period of time, social media did produce better outcomes for revolutions and mass movements.
The researchers, who published their findings in a recent issue of Political Science Research and Methods, specifically examined social media from both the Venezuela regime and its opposition.
Following the death of Venezuelan President Hugo Chavez in early 2013, Nicolas Maduro, Chavez’s vice-president, won a special election.
After his election, mass protests erupted related to economic decline and increased crime.
In their analysis, the researchers noted that the regime abruptly shifted its Twitter strategy after protests swept across the country.
The topics of the regime’s tweets became even more diverse than usual — including such topics as a tree-planting event — and often did not address the protests at all.
As the protests continued, however, the researchers said that the opposition also became less focused, which the researchers suggest may have been a reaction to the regime’s social media strategy.
The way that attention works on social networks offers a glimpse into why the strategy to distract citizens might be effective, added Munger, who worked on the study while a doctoral student in politics at New York University.
“To have effective protests, you need to have a ton of people coordinated on a single message, so spreading other narratives disrupts that process of coordination,” said Munger.
“Being able to spread doubt is effective. You don’t have to get people to love your regime, you just need people to less convinced of the single narrative.”
The regime also seemed to develop a more sophisticated approach to using hashtags. The regime used long hashtags, as opposed to the shorter hashtags that are more commonly used, to promote distraction among the protest groups. (IANS)