Researchers from the University of Tsukuba in Japan have developed an artificial intelligence-powered cryptocurrency portfolio management system called CryptoRLPM. This system is the first of its kind to utilize on-chain data for training, according to the scientists.
CryptoRLPM, short for “Cryptocurrency reinforcement learning portfolio manager,” utilizes a training technique known as reinforcement learning (RL) to implement on-chain data into its model. RL is an optimization paradigm wherein an AI system interacts with its environment, in this case, a cryptocurrency portfolio, and updates its training based on reward signals.
The researchers applied feedback from RL throughout CryptoRLPM’s architecture. The system is structured into five primary units that work together to process information and manage structured portfolios. These modules include a data feed unit, data refinement unit, portfolio agent unit, live trading unit, and agent updating unit.
To test the effectiveness of CryptoRLPM, the scientists assigned it three different portfolios. The first portfolio consisted of only Bitcoin (BTC) and Storj (STORJ), the second portfolio added Bluzelle (BLZ) to BTC and STORJ, and the third portfolio included Chainlink (LINK) alongside the other three assets.
The experiments were conducted over a two-year period from October 2020 to September 2022, and three distinct phases were carried out: training, validation, and backtesting.
The researchers evaluated the success of CryptoRLPM by comparing its performance against a baseline evaluation of standard market performance using three metrics: accumulated rate of return (AAR), daily rate of return (DRR), and Sortino ratio (SR). AAR and DRR measure the amount of gain or loss an asset experienced within a given time period, while SR measures an asset’s risk-adjusted return.
According to the researchers’ pre-print research paper, CryptoRLPM demonstrated significant improvements over the baseline performance for Bitcoin. Specifically, it showed at least an 83.14% improvement in AAR, at least a 0.5603% improvement in DRR, and at least a 2.1767 improvement in SR.
This research explores the potential of AI-powered systems in managing cryptocurrency portfolios. With the ability to utilize on-chain data and reinforcement learning, CryptoRLPM shows promise in optimizing portfolio management by adapting and learning from market conditions. These findings contribute to the growing interest in the intersection of AI and cryptocurrency, with the potential for further advancements in the field.
Source link
