Computational modeling study reveals effective effort allocation between investment and labor on tokenized platforms, highlighting enhanced participant earnings under market uncertainty.
Key Points
To develop an optimal decision-making framework that balances investment and labor intensities for participants seeking to maximize earnings on tokenized blockchain platforms.
Separated individual participant actions from platform averages across two subproblems: metric-based heuristics using Monte Carlo ensembles (assuming negligible individual impact) and a Markov decision process solved with reinforcement learning (modeling explicit individual impact).
Incorporated parameter uncertainty from model estimation, system uncertainty from projections, and input uncertainty from participant-platform interactions.
Evaluated the proposed strategies using empirical historical token price time series.
Both metric-based Monte Carlo projections and reinforcement learning strategies demonstrated strong performance in optimizing overall participant revenue across historical token price series.
The reinforcement learning framework effectively adapted to dynamic state changes when participant interactions exerted direct influence on the platform state.