🤖 AI Summary
Pricing and hedging option portfolios under complex stochastic volatility models (e.g., Heston, SABR) suffer from low computational efficiency, poor robustness, and slow adaptation to changing market conditions.
Method: We propose a deep hedging framework based on low-dimensional model embedding. First, a neural network learns a universal hedge policy representation across diverse model paths. Second, an embedding-driven fast recalibration mechanism enables efficient parameter updates without full network retraining. Third, we systematically characterize the PnL distribution over the model parameter space—enabling model-agnostic hedge construction.
Results: Experiments demonstrate substantial improvements in training efficiency and cross-model/market-state generalization. The framework achieves robust PnL performance on Monte Carlo–generated complex synthetic data, providing a scalable simulation foundation and practical paradigm for model-agnostic hedging.
📝 Abstract
We present a robust Deep Hedging framework for the pricing and hedging of option portfolios that significantly improves training efficiency and model robustness. In particular, we propose a neural model for training model embeddings which utilizes the paths of several advanced equity option models with stochastic volatility in order to learn the relationships that exist between hedging strategies. A key advantage of the proposed method is its ability to rapidly and reliably adapt to new market regimes through the recalibration of a low-dimensional embedding vector, rather than retraining the entire network. Moreover, we examine the observed Profit and Loss distributions on the parameter space of the models used to learn the embeddings. The results show that the proposed framework works well with data generated by complex models and can serve as a construction basis for an efficient and robust simulation tool for the systematic development of an entirely model-independent hedging strategy.