Reinforcement learning and artificial agency
Bibliographic Data
| ID | 21501767 |
|---|---|
| Authors | Patrick Butlin (0000-0001-5837-5057, Future of Humanity Institute University of Oxford Oxford UK, corresponding author) |
| Year | 2024 |
| Volume | 39 |
| Issue | 1 |
| Pages | 22-38 |
| Publication date | 2024-02-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Mind & Language (JOURNAL) |
| Journal identifiers | ISSN: 0268-1064 • E-ISSN: 1468-0017 |
| Publisher | Wiley (PUBLISHER • GB) |
| DOI | 10.1111/mila.12458 |
| OpenAlex | W4382762555 |
| Language | EN |
| Citations received | 10 |
| References cited | 51 |
There is an apparent connection between reinforcement learning and agency. Artificial entities controlled by reinforcement learning algorithms are standardly referred to as agents, and the mainstream view in the psychology and neuroscience of agency is that humans and other animals are reinforcement learners. This article examines this connection, focusing on artificial reinforcement learning systems and assuming that there are various forms of agency. Artificial reinforcement learning systems satisfy plausible conditions for minimal agency, and those which use models of the environment to perform forward search are capable of a form of agency which may reasonably be called action for reasons
Cognitive psychology · Cognitive science · Epistemology · Mainstream · Reinforcement · Reinforcement learning · Computer Science · Evolutionary Game Theory and Cooperation · Free Will and Agency · Law · Psychology · Reinforcement Learning in Robotics · Social Psychology · Artificial Intelligence
Constructing AI power
I gotta use words when I talk to you
Preference‐Based Representations for Collective Agency
Moral Agency without Consciousness
AI welfare risks
Values in science and AI alignment research
The agency in language agents
Current cases of AI misalignment and their implications for future risks
What Is It for a Machine Learning Model to Have a Capability
Understanding Artificial Agency
Language, Thought, and Other Biological Categories
Representation in Cognitive Science
What Biological Functions Are and Why They Matter
How to Count Animals, more or less
Explaining Behavior
Confusion of Tongues
The Importance of Being Rational
Human-level control through deep reinforcement learning
A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
Defining Agency
Goals and Habits in the Brain
ImageNet classification with deep convolutional neural networks
Mastering the game of Go with deep neural networks and tree search
A Neural Substrate of Prediction and Reward
Insightful artificial intelligence
Desire and What It’s Rational to Do
Primitive Agency and Natural Norms
An Organizational Account of Biological Functions
Determined by Reasons
Fellow Creatures
Reinforcement learning
| Unique citing works | 10 |
|---|---|
| Citations per year | 3,33 |
| Citation span | 2023 - 2026 (4) |
| Citation velocity | current |
| Highly cited | No |
| Citation types | Neutral: 9 |