Artificial Intelligence has been the hot topic at CES 2016, Davos World Economic Forum this past week, and today, Hong Kong based Ben Goertzel, world expert on AI and Founder of Aidyia, talked about his hedge fund’s trading algorithms that go beyond the technical trading patterns formulated from price data.
Goertzel’s AI will perform predictive models of price movements without human actions at the computer keys. According to Wired today, he stated, “If we all die, it would keep trading.” The AI driven hedge fund made 2 percent profit its first trading day. This is not a large monetary amount but there are other funds world-wide looking at this technological change in financial trading.
The San Francisco startup Sentient Technologies has been using an AI system since last year. Data driven funds using AI include Bridgewater, AQR Capital and Renaissance Technologies. Other Wall Street names Ray Dalio and Steven A. Cohen are preparing AI trading in their Point72 Asset Management fund.
The use of AI finds itself at Harvard University where $28 million has been granted from Advanced Projects Activity (IARPA) to investigate why brains are better at learning and retaining information than artificial intelligence. If the study can penetrate the variables of the process, than AI can become faster, smarter and function as the human brain.
The human brain can look at an object such as a car and recognize it, but the AI system must process hundreds or thousands of samples to process it and retina the information. AI, however, can process data faster than the human brain.
Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS), Centre for Brain Studies (CBS) and the Department of Molecular and Cellular Biologies will study activity inside the visual cortex. Once, they understand how neurons are connected to each other they will be able to create a more accurate and complex artificial intelligence system. The next step will be the first computer systems that can interpret analyses and learn information as quickly and successfully as human beings.
There is a technique called “deep learning.” This approach uses a technology to recognize images, identify spoken words, and perform other tasks inside Internet companies like Google and Microsoft. This would allow a system to recognize changes in the market which a quantitative based model can’t perform.
Goertzel looks at artificial intelligence and cautions, “If everyone is using something, its predictions will be priced into the market.” He finds that words and pictures do not appear in the financial markets so this approach will not work in AI for trading.
There are 1,360 hedge funds and the use of computer models guides the trades. Goertzel believes that the world of finance and trading is not just about being smart, “but from being smart in a different way from others.”