:warning: About math expressions in the paper

[tolgacangoz]

Thanks for this great survey! I have several questions:

1. I might be wrong: shouldn't s=0 be s=1 at the top of the 3rd equation?
   

2. Shouldn't s𝜃(x, 𝑇), s𝜃(x, 𝑇−1), · · · , s𝜃(x, 0) be s𝜃(x, 𝑇), s𝜃(x, 𝑇−1), · · · , s𝜃(x, 1) in the last paragraph of 8th page?

3. Shouldn't dx = 𝑓 (x𝑡 , 𝑡) − 𝑔(𝑡) 2 s𝜃 (x𝑡 , 𝑡)d𝑡 + 𝑔(𝑡)dw. be dx = [𝑓 (x𝑡 , 𝑡) − 𝑔(𝑡) 2 s𝜃 (x𝑡 , 𝑡)]d𝑡 + 𝑔(𝑡)dw. in the 34th equation?

4. Shouldn't max(𝜖𝑎𝑏𝑠, 𝜖𝑟𝑒𝑙 max(| xlow, | xprev |)) be max(𝜖𝑎𝑏𝑠, 𝜖𝑟𝑒𝑙 max(| xlow |, | xprev |)) at the below of 24th equation?

5. "simply" -> "simplify" under the 2nd equation.

6. "pre-traied" -> "pre-trained" in the last paragraph of the 15th page.

@YangLing0818