mlsanityData to 14 Oct 2026

Plasticity as the Mirror of Empowerment

arXiv 2505.10361 ↗First posted 15 May 2025cs.AI cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

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

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 19 of its 56 references that are in the map. First posted on arXiv 15 May 2025.

Builds on

19 of 56 references.

References · 19

  1. 2016Variational Intrinsic Controlcited by 267
  2. 2023Understanding plasticity in neural networkscited by 174
  3. 2015Variational Information Maximisation for Intrinsically Motivated Reinforcement Learningcited by 144
  4. 2020Explore, Discover and Learn: Unsupervised Discovery of State-Covering Skillscited by 122
  5. 2023Loss of Plasticity in Continual Deep Reinforcement Learningcited by 121
  6. 2019Optimal Policies Tend to Seek Powercited by 119
  7. 2023A Definition of Continual Reinforcement Learningcited by 101
  8. 2021Continual Backprop: Stochastic Gradient Descent with Persistent Randomnesscited by 80
  9. 2021Reinforcement Learning, Bit by Bitcited by 59
  10. 2021Simple Agent, Complex Environment: Efficient Reinforcement Learning with Agent Statescited by 43
  11. 2023Continual Learning as Computationally Constrained Reinforcement Learningcited by 37
  12. 2020Relative Variational Intrinsic Controlcited by 34
7 more
  1. 2022Settling the Reward Hypothesiscited by 33
  2. 2025Characterizing AI Agents for Alignment and Governancecited by 30
  3. 2021Formalizing the Generalization-Forgetting Trade-off in Continual Learningcited by 24
  4. 2025Rethinking the Foundations for Continual Reinforcement Learningcited by 15
  5. 2024Three Dogmas of Reinforcement Learningcited by 14
  6. 2023Hierarchical Empowerment: Towards Tractable Empowerment-Based Skill Learningcited by 7
  7. 2023On the Convergence of Bounded Agentscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 10.

12 papers.