Satisfied, Not Solving: Wireheading of Bounded Homeostatic Rewards in Reinforcement Learning
cs.LG, cs.AI
Submitted: 2026-10-31
Updated: 2026-10-31
Code: https://github.com/chasemetoyer/satisfied-not-solving
Terminology
Sources
- Does Self-Evaluation Enable Wireheading in Language Models?
- Concrete Problems in AI Safety
- Emergence of Addictive Behaviors in Reinforcement Learning Agents
- Unifying Count-Based Exploration and Intrinsic Motivation
- Large-Scale Study of Curiosity-Driven Learning
- Exploration by Random Network Distillation
- Reward Hacking in Language Model Agents: Revisiting AI Safety Gridworlds
- Minigrid & Miniworld: Modular & Customizable Reinforcement Learning Environments for Goal-Oriented Tasks
- Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models
- Avoiding Wireheading with Value Reinforcement Learning
- Reward Tampering Problems and Solutions in Reinforcement Learning: A Causal Influence Diagram Perspective
- CleanRL: High-quality Single-file Implementations of Deep Reinforcement Learning Algorithms
- Categorizing Wireheading in Partially Embedded Agents
- Specification Gaming and a Constrained Reinforcement Learning Mitigation in a Cleaning Gridworld
- The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
- Curiosity-driven Exploration by Self-supervised Prediction
- From homeostasis to resource sharing: Biologically and economically aligned multi-objective multi-agent gridworld-based AI safety benchmarks
- RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis