Hedge funds are training computers to think like traders with an artificial intelligence technology that is reportedly holding out promise for firms. The firm is looking at A.I., a kind of machine learning on steroids, to serve as an edge in the global finance.

Nicolas Chapados, a computer scientist, said the technology is about five years away from becoming a mainstream tool at hedge funds. He notes that there is a huge class of deep-learning models being utilised in tech firms that can be adapted to financial processing. Chapados is one of the co-heads of quant fund Chapados Couture Capital in Montreal and Element AI.

However, researchers are cautious to not oversell the technology after previous round of disappointment. Winton, a quant firm in London, had released a statement, saying they tend to be cynical of claims that deep learning will solve the general problem of investment management.

A 2015 study of deep learning’s ability to forecast moves of 15 stocks in the S&P 500 Index showed that over 10 million events were pulled from financial news as part of the training of the computer system. A team of researchers led by Xiao Ding of the Harbin Institute of Technology’s Research Center for Social Computing and Information Retrieval in China said the model made a cumulative profit in simulating trading of the stocks.

Chapados had explained that the neural net approach will begin with a model, which has to be flexible so it can represent comprehensive classed on possible non linearity instead of investing manpower to discover what non linearities are relevant. There would be no need to test out every single one of them to see if they are significant.

Hitoshi Harada, co-founder of Alpaca, a startup that sells deep learning trading platforms, recognised that deep learning techniques can work well with the use of millions or billions of data. However, he said it would be challenging to build a good model with only a thousand data points of single-stock data per day, citing companies have publicly traded for only a year.

“We do, however, believe that technology has advanced sufficiently over recent years to suggest such methods have the potential to improve elements of our process, and deep learning is an active area of long-term research for Winton," the hedge fund said according to Bloomberg. Despite the challenges, Winton researchers pursue the deep learning as a useful weapon.