Philosophy Dictionary of ArgumentsHome
| |||
|
| |||
| Reinforcement learning: Reinforcement learning (RL) is a type of machine learning that allows an agent to learn how to behave in an environment by trial and error. See also Learning, Machine learning, Artificial Intelligence._____________Annotation: The above characterizations of concepts are neither definitions nor exhausting presentations of problems related to them. Instead, they are intended to give a short introduction to the contributions below. – Lexicon of Arguments. | |||
| Author | Concept | Summary/Quotes | Sources |
|---|---|---|---|
|
Tom Griffiths on Reinforcement Learning - Dictionary of Arguments
Brockman I 127 Reinforcement Learning/Griffiths: Reinforcement learning is a standard method for training intelligent machines. By associating particular outcomes with rewards, a machine-learning system Brockman I 128 can be trained to follow strategies that produce those outcomes. Def Inverse reinforcement learning: turns this approach around: By observing the actions of an intelligent agent that has already learned effective strategies, we can infer the rewards that led to the development of those strategies. Brockman I 130 Rationality is the standard assumption in inverse-reinforcement-learning models that try to make inferences from human behavior - perhaps with the concession that humans are not perfectly rational agents and sometimes randomly choose to act in ways unaligned with or even opposed to their best interests. Brockman I 131 (…) heuristic is a reasonable strategy for avoiding complex probabilistic computations, but also results in errors. For instance, relying on the ease of generating an event from memory as a guide to its probability leads us to overestimate the chances of extreme (hence extremely memorable) events such as terrorist attacks. Heuristics provide a more accurate model of human cognition but one that is not easily generalizable. How do we know which heuristic people might use in a particular situation? Are there other heuristics they use that we just haven’t discovered yet? >Decision Theory/Griffiths. Griffiths, Tom, “The Artificial Use of Human Beings” in: Brockman, John (ed.) 2019. Twenty-Five Ways of Looking at AI. New York: Penguin Press._____________Explanation of symbols: Roman numerals indicate the source, arabic numerals indicate the page number. The corresponding books are indicated on the right hand side. ((s)…): Comment by the sender of the contribution. Translations: Dictionary of Arguments The note [Concept/Author], [Author1]Vs[Author2] or [Author]Vs[term] resp. "problem:"/"solution:", "old:"/"new:" and "thesis:" is an addition from the Dictionary of Arguments. If a German edition is specified, the page numbers refer to this edition. |
Griffiths, Tom Brockman I John Brockman Possible Minds: Twenty-Five Ways of Looking at AI New York 2019 |
||
> Counter arguments against Griffiths
> Counter arguments in relation to Reinforcement Learning
Ed. Martin Schulz, access date 2024-04-27