Research map / Reinforcement learning / Imitation learning
Rationally Inattentive Inverse Reinforcement Learning Explains YouTube Commenting Behavior
William Hoiles, Vikram Krishnamurthy, Kunal Pattanayak
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| drop the rational inattention cost constraint from the utility model (ablation) | Rationally inattentive utility model · YouTube behavior prediction (Hamming distance) | mean squared prediction error: 3.85 -> 2.45 | Improved |
| increase Renyi parameter beta (Renyi mutual information cost) (ablation) | Renyi cost model · YouTube categories | category pairs passing test: 0 -> 2 -> 3 pairs | Improved |
| use Shannon mutual information cost instead of general cost (ablation) | Rational inattention test · YouTube category pairs (18 choose 2) | pairs passing test: 45 -> 2 | Hurt |
| increase beta in Renyi cost (ablation) | Renyi cost model · YouTube robustness R3 | average R3: average R3 <= 0.01 for beta >= 0.9 | Improved |
Closest papers by meaning
- Quantifying the Sensitivity of Inverse Reinforcement Learning to Misspecification (ICLR 2024 · similarity 0.86)
- Imitating Language via Scalable Inverse Reinforcement Learning (NeurIPS 2024 · similarity 0.85)
- InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modeling (NeurIPS 2024 · similarity 0.84)
- Distributional Preference Learning: Understanding and Accounting for Hidden Context in RLHF (ICLR 2024 · similarity 0.84)
- MTRec: Learning to Align with User Preferences via Mental Reward Models (NeurIPS 2025 · similarity 0.84)
- Capturing Individual Human Preferences with Reward Features (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
