id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_43tkopwxz5giffwdwgogqhn3xq	M.A.H. Dempster	An automated FX trading system using adaptive reinforcement learning	2006	21	.pdf	application/pdf	8903	3285	91	relies on a layered structure consisting of a machine learning algorithm, a risk management manages risk in an intelligent manner so that it protects past gains and avoids losses by restraining or even shutting down trading activity in times of high uncertainty. dynamically optimizes the global trading performance in terms of a trader's risk preferences by automatically tuning the system's hyper-parameters. machine-learning algorithm, the risk and performance management layer and the dynamic optimization the parameter δ will influence the performance and risk profile of the resulting trading We also improved the performance of the RRL trading model by implementing a better position updating scheme. A higher risk aversion parameter ν induces the optimization to find a performance profile trading system becomes less risky as the risk aversion parameter in the layer 3 optimization is increased. risk aversions of the trading system with and without Layer 3 activated.	./cache/work_43tkopwxz5giffwdwgogqhn3xq.pdf	./txt/work_43tkopwxz5giffwdwgogqhn3xq.txt