How quantitative funds use machine learning systems like aidyia to invest in the stock market
Aidyia, a Hong Kong automated system, uses a number of Artificial Intelligence (AI) capabilities, including one inspired by genetic evolution and another one by probabilistic logic, to make predictions about the market. On its first day, it generated a 2% increase in an undisclosed amount of money, according to Ben Goertzel, its creator and a long-time AI guru. Every day these engines process vast quantities of material–analysing everything from market prices to macroeconomic data and corporate accounting documents–and then “vote” on a course of action. According to a study by Preqin, a market research firm, some 1,360 quantitative funds use computer models to trade. This is 9% of all funds, and it represents $197 billion in total. Below are three reasons why the AI trend could alter the dynamics of markets.
A Resurgence of AI Technology
New machine learning techniques, especially deep learning, have allowed computers to recognise materials such as images, text, and audio. Many tech companies, quantitative funds and financial firms are exploring whether methods like deep learning lend themselves to finance. Furthermore, new AI techniques can now use unstructured natural language input like news articles, company reports, and social media posts to guide their decisions.
Evolutionary Computation Systems Learn from Their Mistakes
Systems like Aidyia create a large number of random digital stock traders and test their performance with historical stock data. Then the system selects the winners and keeps their “genes” to create a new generation of superior traders. This process repeats indefinitely, and over thousands of generations, trillions of individuals compete. Eventually, the system produces a smart trader population.
The Use of a Wide Range of Technologies
If a trick that works is found, sceptics point to the likelihood that other funds and investors will latch on to it and pour in their money. In this way, any advantage will be arbitraged away. But Goertzel sees this risk and says that his big idea is competition. To do something that no human or machine is doing, and for this, he believes that a wide range of technologies and machine learning are critical.