Hi Arvind, thanks for the going down the hillside analogy for gradient descent. Please check this set of posts for what I hope is a simple explanation for the ideal learning rate = inverse of the second derivative.

https://medium.com/@ranjeettate/learning-rate-in-gradient-descent-and-second-derivatives-2-3c2f8b837ad8