Why reward is not enough for intelligence
In an essay for TechTalks, Herbert Roitblat, data scientist and author of Algorithms Are Not Enough, explains why rewards are not enough to build artificial intelligence.
Roitblat’s essay is in response to a recent paper by DeepMind scientists, in which they claim that rewards and reinforcement learning are all you need to build all the components of intelligence.
In his essay, Roitblat draws lessons from decades of work on cognitive science and artificial intelligence to refute the claims made by DeepMind.
For more on the topic: