Plasticity as the Mirror of Empowerment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Abel, David, Bowling, Michael, Barreto, André, Dabney, Will, Dong, Shi, Hansen, Steven, Harutyunyan, Anna, Khetarpal, Khimya, Lyle, Clare, Pascanu, Razvan, Piliouras, Georgios, Precup, Doina, Richens, Jonathan, Rowland, Mark, Schaul, Tom, Singh, Satinder
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908619443798016
author Abel, David
Bowling, Michael
Barreto, André
Dabney, Will
Dong, Shi
Hansen, Steven
Harutyunyan, Anna
Khetarpal, Khimya
Lyle, Clare
Pascanu, Razvan
Piliouras, Georgios
Precup, Doina
Richens, Jonathan
Rowland, Mark
Schaul, Tom
Singh, Satinder
author_facet Abel, David
Bowling, Michael
Barreto, André
Dabney, Will
Dong, Shi
Hansen, Steven
Harutyunyan, Anna
Khetarpal, Khimya
Lyle, Clare
Pascanu, Razvan
Piliouras, Georgios
Precup, Doina
Richens, Jonathan
Rowland, Mark
Schaul, Tom
Singh, Satinder
contents Agents are minimally entities that are influenced by their past observations and act to influence future observations. This latter capacity is captured by empowerment, which has served as a vital framing concept across artificial intelligence and cognitive science. This former capacity, however, is equally foundational: In what ways, and to what extent, can an agent be influenced by what it observes? In this paper, we ground this concept in a universal agent-centric measure that we refer to as plasticity, and reveal a fundamental connection to empowerment. Following a set of desiderata on a suitable definition, we define plasticity using a new information-theoretic quantity we call the generalized directed information. We show that this new quantity strictly generalizes the directed information introduced by Massey (1990) while preserving all of its desirable properties. Under this definition, we find that plasticity is well thought of as the mirror of empowerment: The two concepts are defined using the same measure, with only the direction of influence reversed. Our main result establishes a tension between the plasticity and empowerment of an agent, suggesting that agent design needs to be mindful of both characteristics. We explore the implications of these findings, and suggest that plasticity, empowerment, and their relationship are essential to understanding agency
format Preprint
id arxiv_https___arxiv_org_abs_2505_10361
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Plasticity as the Mirror of Empowerment
Abel, David
Bowling, Michael
Barreto, André
Dabney, Will
Dong, Shi
Hansen, Steven
Harutyunyan, Anna
Khetarpal, Khimya
Lyle, Clare
Pascanu, Razvan
Piliouras, Georgios
Precup, Doina
Richens, Jonathan
Rowland, Mark
Schaul, Tom
Singh, Satinder
Artificial Intelligence
Machine Learning
Agents are minimally entities that are influenced by their past observations and act to influence future observations. This latter capacity is captured by empowerment, which has served as a vital framing concept across artificial intelligence and cognitive science. This former capacity, however, is equally foundational: In what ways, and to what extent, can an agent be influenced by what it observes? In this paper, we ground this concept in a universal agent-centric measure that we refer to as plasticity, and reveal a fundamental connection to empowerment. Following a set of desiderata on a suitable definition, we define plasticity using a new information-theoretic quantity we call the generalized directed information. We show that this new quantity strictly generalizes the directed information introduced by Massey (1990) while preserving all of its desirable properties. Under this definition, we find that plasticity is well thought of as the mirror of empowerment: The two concepts are defined using the same measure, with only the direction of influence reversed. Our main result establishes a tension between the plasticity and empowerment of an agent, suggesting that agent design needs to be mindful of both characteristics. We explore the implications of these findings, and suggest that plasticity, empowerment, and their relationship are essential to understanding agency
title Plasticity as the Mirror of Empowerment
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2505.10361