backtrader策略库：强化学习一: 梯度提升（ Gradient Ascent）

本文来自博客文章，文末含源码链接。

In the next few posts, I will be going over a strategy that uses Machine Learning to determine what trades to execute. Before we start going over the strategy, we will go over one of the algorithms it uses: Gradient Ascent.

What is Gradient Ascent?

Gradient ascent is an algorithm used to maximize a given reward function. A common method to describe gradient ascent uses the following scenario: Imagine you are blindfolded and placed somewhere on a mountain. Your task is then to find the highest point of the mountain. In this scenario, the “reward function” you are trying to maximize is your elevation. A simple method of locating this maximum would be observing the slope of the area you are standing on, and moving uphill. Following these directions one step a