• A symbolic regression model predicts corrosion rates of biodegradable Zn–0.45Mn–0.2Mg alloy with a determination coefficient of 0.97, outperforming five other ML models.
• Prediction errors in verification experiments were less than 10%, demonstrating high accuracy and reliability.
• The study provides an explicit analytical equation linking corrosion rate to four corrosion parameters, enabling quantitative analysis.
• This data-driven approach combined with accelerated corrosion testing offers a promising strategy for biodegradable metal implant research.