Fruit Fly Brain Simulation Makes $1 Profit in First Day of Bitcoin Trading
A digital simulation of a fruit fly's neural network made $1 in profit after one day of trading Bitcoin on Coinbase, but its creator says this is not enough to prove its investment learning ability.
The experiment, called Stonkfly, was funded by software engineer Alex Warmus with an initial $100. It simulates the connectivity map of a fruit fly's central nervous system and links it to cryptocurrency price data from Coinbase.
The trading process is unusual, as it takes public market data and converts it into RGB images, which are then fed into thousands of sensory inputs corresponding to neurons in an insect visual system. This generates buy and sell signals based on profit and loss.
When the portfolio makes a profit, 15 dopamine neurons are stimulated, while when a loss occurs, 2 avoidance response neurons are activated. However, as Warmus notes, a single day's profit does not necessarily prove that Stonkfly has learned market patterns or can deliver repeatable investment results.