id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-146214-lp78l776	Leal, Laura	Learning a functional control for high-frequency finance	2020-06-17		.txt	text/plain	7922	377	61	Moreover, to answer to genuine requests of financial regulators on the explainability of machine learning generated controls, we project the obtained"blackbox controls"on the space usually spanned by the closed-form solution of the stylized optimal trading problem, leading to a transparent structure. Our paper addresses this last case, belonging to the academic field of optimal trading, initially introduced by [1] and [3] , and then extended in a lot of ways, from sophisticated stochastic control [5] to Gaussianquadratic approximations allowing to obtain closed-form solutions like [13] or [14] , or under a self-financing equation context in [12] . We start by the functional space of controls spanned by the closed-form solution of the stylized problem: they are non-linear in the remaining time to trade and affine in the remaining quantity to trade (see [9] for a description of the relationship between the optimal controls and the space generated by the h 1 (t) and h 2 (t) defined later in the paper).	./cache/cord-146214-lp78l776.txt	./txt/cord-146214-lp78l776.txt