: Reinforcement Learning Trading Agent in a Rough Volatility Setting

Since financial markets are constantly changing, stochastic control techniques are implemented to design trading strategies. The amount of market data and unstructured data is rapidly increasing. In order to use as much as possible of the data in trading algorithms, complex computer algorithms are needed. For these computer algorithms to have the desired outcome and effect, it is vital that they are trusted by the financial institutions using them, by the regulators, and by society in general. As a result of the non-stationarity of financial time series, the computer algorithms need to be resilient to change and capable of functioning amid such change whilst also managing risk. It is also important to embed a sense of accountability and responsibility both in the systems and the people designing such systems. Since the stability of the financial markets is crucial to society and the economy, researching trading systems with a focus on their trustworthiness is crucial.