17.2 C
London
Thursday, 28 September, 2023

Videos from CppCon 2021

CppCon is the annual, week-long online face-to-face...

Super Fast Python

A new website by Jason Brownlee aims...

Machine Learning in Cardiovascular Medicine

The book Machine Learning in Cardiovascular Medicine...

Reward is enough

AIReward is enough

In a recent paper by David Silver, Satinder Baveja, Doina Precup, and Richard Sutton, the authors hypothesize that the objective of maximizing reward is enough to drive behaviour that exhibits most if not all attributes of intelligence that are studied in natural and artificial intelligence, including knowledge, learning, perception, social intelligence, language, and generalization. This is in contrast to the view that specialized problem formulations are needed for each attribute of intelligence, based on other signals or objectives. The reward-is-enough hypothesis suggests that agents with powerful reinforcement learning algorithms when placed in rich environments with simple rewards could develop the kind of broad, multi-attribute intelligence that constitutes an artificial general intelligence.

Check out our other content

Videos from CppCon 2021

Super Fast Python

Check out other tags:

Most Popular Articles