Emotion-Gradient Metacognitive RSI (Part I): Theoretical Foundations and Single-Agent Architecture
On the map: Deep reinforcement learning methods → RL training of language and agent models
Abstract (arXiv)
We present the Emotion-Gradient Metacognitive Recursive Self-Improvement (EG-MRSI) framework, a novel architecture that integrates introspective metacognition, emotion-based intrinsic motivation, and recursive self-modification into a unified theoretical system. The framework is explicitly capable of overwriting its own learning algorithm under formally bounded risk. Building upon the Noise-to-Meaning RSI (N2M-RSI) foundation, EG-MRSI introduces a differentiable intrinsic reward function driven by confidence, error, novelty, and cumulative success. This signal regulates both a metacognitive mapping and a self-modification operator constrained by provable safety mechanisms. We formally define the initial agent configuration, emotion-gradient dynamics, and RSI trigger conditions, and derive a reinforcement-compatible optimization objective that guides the agent's development trajectory. Meaning Density and Meaning Conversion Efficiency are introduced as quantifiable metrics of semantic learning, closing the gap between internal structure and predictive informativeness. This Part I paper establishes the single-agent theoretical foundations of EG-MRSI. Future parts will extend this framework to include safety certificates and rollback protocols (Part II), collective intelligence mechanisms (Part III), and feasibility constraints including thermodynamic and computational limits (Part IV). Together, the EG-MRSI series provides a rigorous, extensible foundation for open-ended and safe AGI.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 4 of its 20 references that are in the map. First posted on arXiv 12 May 2025.
Builds on
4 of 20 references.
References · 4
- 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
- 2017Emotion in Reinforcement Learning Agents and Robots: A Surveycited by 21
- 2015From Seed AI to Technological Singularity via Recursively Self-Improving Softwarecited by 9
- 2025When Your Own Output Becomes Your Training Data: Noise-to-Meaning Loops and a Formal RSI Triggercited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Truly Self-Improving Agents Require Intrinsic Metacognitive Learning
- 2025Emotion-Inspired Learning Signals (EILS): A Homeostatic Framework for Adaptive Autonomous Agents
- 2026Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs
- 2026From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement
- 2026MIRROR: A Hierarchical Benchmark for Metacognitive Calibration in Large Language Models
- 2026Learn Like Humans: Use Meta-cognitive Reflection for Efficient Self-Improvement
- 2026Meta-Learning and Meta-Reinforcement Learning -- Tracing the Path towards DeepMind's Adaptive Agent
- 2026MA2P: A Meta-Cognitive Autonomous Intelligent Agents Framework for Complex Persuasion