Ex. 2: Incorrect torque reward formulation?

[rabaur]

I think the formulation given in the README.md for the torque reward in exercise 2 is wrong. The weight term should be outside the exponential. Before doing so, the tests fail (but moving it out works)

[MiguelZamoraM]

That is a typo on the README. Indeed the right expression for torque is: $w_{\tau} \text{exp}(-\Vert \mathbf{\tau} \Vert_2^2 / (2 N \sigma_t^2))$

I just fixed it on the README. Thanks for mentioning it!