In a new paper, scientists at Google’s DeepMind AI research unit argue that intelligence and its associated abilities will emerge not from formulating and solving complicated problems but by sticking to a simple but powerful principle: reward maximization.
Reinforcement learning formalises the problem of goal-seeking intelligence. The general problem may be instantiated with a wide and realistic range of goals and worlds, corresponding to different reward signals to maximise in different environments.
The researchers also acknowledge that learning mechanisms for reward maximization is an unsolved problem that remains a central question to be further studied in reinforcement learning.
MORE