Intervention Complexity as a Canonical Reward and a Measure of Intelligence

• URL: http://arxiv.org/abs/2605.02175v2
• Date: Sat, 09 May 2026 02:59:58 GMT
• Title: Intervention Complexity as a Canonical Reward and a Measure of Intelligence
• Abstract summary: We propose a new measure, called intervention complexity, that has five natural properties: environment-derivedness, universality, minimality, sensitivity, and achievement preference.<n>The result yields a family of canonical rewards indexed by resource bias.<n>We discuss implications for superintelligence and for pre-training universal agents.
• Score: 1.2183405753834562
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: The Legg--Hutter universal intelligence measure provides a rigorous scalar assessment of general intelligence as expected reward across all computable environments, weighted by simplicity. However, the measure presupposes an externally specified reward function, raising the question of whether the reward primitive is inherently arbitrary or whether a canonical choice exists. We propose a new measure, called intervention complexity, that has five natural properties: environment-derivedness, universality, minimality, sensitivity, and achievement preference. Given a resource function rho encoding an inductive bias (such as program length, execution time, or energy), rho-intervention complexity is a universal reward. The result yields a family of canonical rewards indexed by resource bias, providing a principled completion of the Legg--Hutter framework that does not require external normative input. We further propose a two-dimensional characterisation of intelligence: agent competence (how well the agent performs relative to the oracle optimum) and learning efficiency (how quickly this competence improves with experience). A separation theorem establishes that the choice of resource bias determines the computability of the resulting measure: action-count IC is computable in polynomial time, while program-length IC without oracle access is uncomputable, with the gap between oracle and bare IC precisely quantifying the information-theoretic content of learning. We discuss implications for superintelligence and for pre-training universal agents.

Related papers

• CPR-IE:A Compression-Prediction-Resource Intelligence Efficiency Metric [0.685316573653194]
  We develop a protocol-relative ordering by representational economy, predictive quality, and resourceburden.<n>CPR-IE is a constructed efficiency representation, not a universal law or a definition ofintelligence itself.
  arXiv  Detail & Related papers  (2026-09-04T07:06:21Z)
• TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning [52.67756371231985]
  Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models.<n>We introduce Tree Rollout Allocation for Contrastive Exploration (TRACE), a unified rollout allocation framework.<n>Technically, TRACE allocates rollout budget to both prompt roots and intermediate prefixes that are most likely to yield mixed terminal rewards.
  arXiv  Detail & Related papers  (2026-06-09T17:16:03Z)
• Breaking the Computational Barrier: Provably Efficient Actor-Critic for Low-Rank MDPs [53.412166189410904]
  We use supervised learning as a computational proxy to establish a hierarchy of commonly adopted RL oracles under low-rank Markov Decision Processes (MDPs)<n>Motivated by this observation, we propose a novel optimistic actor-critic algorithm that relies solely on the policy evaluation oracle.<n>We prove that our algorithm outperforms the existing sample complexity guarantees for low-rank MDPs while avoiding computationally expensive planning or optimization oracles commonly assumed in prior works.
  arXiv  Detail & Related papers  (2026-05-02T04:46:54Z)
• Provable and Practical In-Context Policy Optimization for Self-Improvement [49.670847804409874]
  We study test-time scaling, where a model improves its answer through multi-round self-reflection at inference.<n>We introduce In-Context Policy Optimization (ICPO), in which an agent optimize its response in context using self-assessed or externally observed rewards without modifying its parameters.<n>We propose Minimum-Entropy ICPO (ME-ICPO), a practical algorithm that iteratively uses its response and self-assessed reward to refine its response in-context at inference time.
  arXiv  Detail & Related papers  (2026-03-02T00:21:50Z)
• From Atomic to Composite: Reinforcement Learning Enables Generalization in Complementary Reasoning [83.94543243783285]
  We study Complementary Reasoning, a complex task that requires integrating internal parametric knowledge with external contextual information.<n>We find that RL acts as a reasoning synthesizer rather than a probability amplifier.
  arXiv  Detail & Related papers  (2025-12-01T18:27:25Z)
• Scalable In-Context Q-Learning [68.9917436397079]
  We propose textbfScalable textbfIn-textbfContext textbfQ-textbfLearning (textbfSICQL) to steer in-context reinforcement learning.<n>textbfSICQL harnesses dynamic programming and world modeling to steer ICRL toward efficient reward and task generalization.
  arXiv  Detail & Related papers  (2025-06-02T04:21:56Z)
• SuperARC: An Agnostic Test for Narrow, General, and Super Intelligence Based On the Principles of Recursive Compression and Algorithmic Probability [0.14061979259370275]
  We introduce an open-ended test grounded in algorithmic probability.<n>It can avoid benchmark contamination in the quantitative evaluation of frontier models.<n>We prove that compression is equivalent and directly proportional to a system's predictive power.
  arXiv  Detail & Related papers  (2025-03-20T23:11:30Z)
• Optimal Query Allocation in Extractive QA with LLMs: A Learning-to-Defer Framework with Theoretical Guarantees [3.4289478404209826]
  Large Language Models excel in generative tasks but exhibit inefficiencies in structured text selection.<n>We propose a Learning-to-Defer framework that allocates queries to specialized experts, ensuring high-confidence predictions.
  arXiv  Detail & Related papers  (2024-10-21T08:21:00Z)
• REBEL: Reward Regularization-Based Approach for Robotic Reinforcement Learning from Human Feedback [61.54791065013767]
  A misalignment between the reward function and human preferences can lead to catastrophic outcomes in the real world.<n>Recent methods aim to mitigate misalignment by learning reward functions from human preferences.<n>We propose a novel concept of reward regularization within the robotic RLHF framework.
  arXiv  Detail & Related papers  (2023-12-22T04:56:37Z)
• Oracle Computability and Turing Reducibility in the Calculus of Inductive Constructions [0.0]
  We develop synthetic notions of oracle computability and Turing reducibility in the Calculus of Inductive Constructions. As usual in synthetic approaches, we employ a definition of oracle computations based on meta-level functions. We show that Turing reducibility forms an upper semilattice, transports decidability, and is strictly more expressive than truth-table reducibility.
  arXiv  Detail & Related papers  (2023-07-28T13:16:46Z)
• Regularized Online Allocation Problems: Fairness and Beyond [7.433931244705934]
  We introduce the emphregularized online allocation problem, a variant that includes a non-linear regularizer acting on the total resource consumption. In this problem, requests repeatedly arrive over time and, for each request, a decision maker needs to take an action that generates a reward and consumes resources. The objective is to simultaneously maximize additively separable rewards and the value of a non-separable regularizer subject to the resource constraints.
  arXiv  Detail & Related papers  (2020-07-01T14:24:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.