Deep Hedging: Continuous Reinforcement Learning for Hedging of General Portfolios across Multiple Risk Aversions

We present a method for finding optimal hedging policies for arbitrary However, since the idealized assumptions of a complete market initial portfolios and market states. We develop a novel actorcritic do not apply in real markets, it is not surprising that complete market algorithm for solving general risk-averse stochastic control models require constant manual adjustments and oversight, for problems and use it to learn hedging strategies across multiple risk example adjusting delta by "skew delta", smoothing barriers priced aversion levels simultaneously. We demonstrate the effectiveness with local volatility, and taking into account market impact when of the approach with a numerical example in a stochastic volatility trading vega to hedge auto-callable products.