Siegel, who has used AI to build his Two Sigma Investments into a $37 billion hedge fund firm, said algorithms are limited by the scant amount of training data available to instruct them on how to differentiate objects in images.
“Artificial intelligence today doesn’t have anything that resembles common sense, and common sense is a key feature of intelligence,” Siegel, 55, said at the Bloomberg Markets Most Influential Summit on Wednesday.
Hedge funds are embracing a form of AI called machine learning years after Two Sigma deployed the technology and as stock and bond pickers struggle to outperform markets. Highbridge Capital Management, Bridgewater Associates and Point72 Asset Management are among firms trying to profit from machine-learning algorithms, which automatically find patterns in large batches of financial data to inform investment calls.
“AI today will solve certain problems but not everything,” Siegel said. “The quest to create general human intelligence, that’s probably a long way in the future.”
Two Sigma, which Siegel co-founded in 2001, is among the success stories in quantitative investing. The New York-based firm’s assets have surged from about $2.5 billion a decade ago. Its Compass fund has returned 12.6 percent this year through August, according to an investor document. That compares with 2.2 percent for hedge funds, according to data compiled by Bloomberg.
Perrin Wheeler, a spokeswoman for the firm, declined to comment on the performance.
Quantitative hedge funds as a group have outperformed traditional managers for years. Quants have gained an average 4.5 percent annually in the past five years, as measured by Hedge Fund Research Inc.’s quantitative directional index. That compares with 3.5 percent for the average hedge fund.
Siegel who earned a Ph.D. in computer science from the Massachusetts Institute of Technology, started Two Sigma with John Overdeck, who had worked at D.E. Shaw & Co. Today, Two Sigma employs more than 1,000 people, including eight gold and silver medalists at the International Math Olympiad, according to its website. The firm processes 22 million gigabytes of data, uses more than 10,000 data sources, has a lab for hackers and hosts robotics competitions. A unit of the firm, called Two Sigma Ventures, seeks to invest in companies focused on data science, machine learning, artificial intelligence and advanced hardware.
Siegel told a conference in May that he’s “very worried” that machines could soon cost large swaths of the global workforce their jobs.
More than a decade after Siegel left his post as chief technology officer at Tudor Investment Corp., the hedge fund firm is adopting quantitative methods. Tudor Investment has recently hired scientists and mathematicians, some with doctorates, to help money managers boost their lackluster performance. Founder Paul Tudor Jones has told colleagues that his firm needs to get up to speed with newer technologies, people with knowledge of the matter said last month.
Investment banks and hedge funds have employed quants to uncover relationships in markets and exploit them with their trades for decades. In the early days, human intervention was needed to teach algorithms how to get smarter. Today’s machine learning is self-taught and is constantly enhancing its intelligence without the need for a helping human hand.
© 2016 Bloomberg L.P
