RL in trading / finance — techniques.
hardAnswer
- (1) Portfolio optimization: policy allocates capital across assets.
- (2) Market making: bid/ask spread policy.
- (3) Options hedging: dynamic hedging via RL.
- Challenges: (1) non-stationary markets.
- (2) Extremely low signal-to-noise.
- (3) Adversarial (other traders learn).
- (4) Backtesting can't capture real market impact.
- Modern practice: careful backtest + shadow trading + tiny position sizes early.
- Most quantitative funds combine RL with classical control + supervised signals.
Check yourself — multiple choice
- Random
- Portfolio allocation / market making / hedging; challenges: non-stationary + low SNR + adversarial + backtest limits; industry combines RL + classical + supervised; shadow trading before live
- Same as poker
- Not real
Finance RL: allocation / MM / hedging; low SNR + adversarial + shadow trading.
#applications
Practise Reinforcement Learning
214 interview questions in this topic.