If you run a policy gradient method on a complex environment and each time you get a different result, it is not weird at all, in fact it is just normal. Please stop expecting them to do what they haven’t promised you, lol 
#DeepRL #nonconvex #stochasticOptimization
Yes, because it is nonconvex. That's why it is important to do many runs, which rarely happens in deep reinforcement learning #deepRL
