How do elite investors use artificial intelligence and machine learning to stand out

Citigroup (C) uses machine learning to make portfolio recommendations to clients. High frequency trading companies use machine learning tools to quickly read and react to financial markets. And stores such as PanAgora Asset Management have developed complex algorithms to test sophisticated investment ideas.

Much of the technology used by the elite investors is not really new. Financial firms are simply better able to leverage the power of AI and machine learning because current computers can process information much faster. And there is a lot more data now than a few years ago.

The rise of machine learning

Nevertheless, technology is rapidly disrupting the financial sector – and will continue to do so.

"The rise of machine learning will make our sector unrecognizable in the future," said Anthony Cowell, head of asset management at KPMG in the Cayman Islands. His clients include some of the largest asset managers, hedge funds and investment capital companies in the world.

Citi Private Bank has, for example, deployed machine learning to help financial advisors answer a frequently asked question: what do other investors do with their money? By using technology, the bank can anonymously share portfolio transactions made by customers across the globe.

"Traditionally, this type of information came from your network.You may have drunk a few cafes or heard about it around a cocktail party," CNSC's Philip Watson, head of the laboratory, told CNN. global investment at Citi and director of innovation at Citi Private Bank. Business. "Now we can share very useful ideas."

Citi has also created a recommendation engine using machine learning tools to advise clients. The platform recommends tailored research reports, solutions and even alerts major event clients such as the maturity of a bond in their portfolio.

Machines help high-speed traders

Domeyard, a Boston hedge fund focused on high-frequency trading, relies on automatic learning to decrypt 300 million data points during the first hour of trading on the New York Stock Exchange.

"We rely on the help of machines to enable us to predict more easily and quickly what will happen in the second or the minute to come," said Christine Qi, co-founder and partner of Domeyard.

But Qi warned that the machines are "as intelligent as the data you provide."

Earlier this year, PanAgora, Boston's Quant Shop, expanded its exposure to China by launching a "self-learning" algorithm that deciphers the Chinese "cyber-slang" used by investors on social media to bypass government censorship, said Chen. The results give PanAgora's portfolio managers a valuable window on the sentiment of particular investors, who dominate the market in China.

Man against machine?

Technology officials have warned not to believe all the hype regarding artificial intelligence and machine learning, especially with regard to the takeover by robots.

"Some effects can be extremely exaggerated," said Watson of Citi. "It's a human world plus machine, it's not just a machine model, I do not see it becoming a machine-only model for a very long time either."

PanAgora's Chen agreed. "It's not the man against the machine – it's the man plus the machine."

At PanAgora, the human being has the last word on investment decisions and sometimes cancels what computer models tell him to do.

"The machines are not sensitive, the terminators will not get up and kill us all in the next 10 years," Chen said. "J & # 39; hope."

Most jobs will be affected

But that does not mean that humans will not be disturbed.

Foster said the most optimistic result is that businesses, governments, and education systems are anticipating this disruption by recycling workers' skills.

"The world is probably moving more slowly than that, there is a risk that people will be left behind," Foster said. "In business, it behooves us to help our workforce to get ahead."

Citi's Citi Watson believes that many workers in lower-level back-office positions could be moved to more rewarding positions.

And after?

"In fact, we do not know it yet," Jackson said at the request of specific use cases. "We are just beginning to understand the power of this."

He said that it was already clear that quantum computers would excel in several areas: encryption, security, chemistry and machine learning.

"He will live up to the hype," Jackson said.

There are still many things that computers can not do in the financial field.

For example, savvy investors often use game theory to map the reaction of other market participants to a given situation. Game theory allows companies to earn money by positioning themselves – before the market experiences strong fluctuations.

PanAgora's Chen said the machines could not do it – for the moment.

"I hope to see it in the next five to fifteen years," he said.