Wojciech Zaremba - Understanding the Reward Function (via searchlore.ai)