Research map / Reinforcement learning / Imitation learning
I See What You See: Inferring Sensor and Policy Models of Human Real-World Motor Behavior
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| infer sensor model with MCE SLQG instead of direct policy estimation (behavior cloning) | driver model · Driver glance prediction | KL divergence: +0.64 -> +0.55 | Helped |
| infer sensor model with MCE SLQG instead of direct policy estimation (behavior cloning) | driver model · Driver state prediction | negative log-likelihood: −7.00 -> −8.58 | Helped |
| infer sensor model parameters jointly with rewards (method 5) instead of best-guess sensor model (method 2) | MCE SLQG · Driver glance prediction | KL divergence: +0.55 -> +0.25 | Helped |
| infer sensor model jointly (method 5) instead of best-guess sensor model (method 2) | MCE SLQG · Driver state prediction | negative log-likelihood: −8.58 -> −8.58 | No clear effect |
| joint estimation (method 5) instead of two-step action-likelihood sensor estimation (method 3) | sensor and policy inference · Driver glance prediction | KL divergence: +0.28 -> +0.25 | Helped |
| MCE-based joint estimation (method 4) instead of method 3 | sensor and policy inference · Driver glance prediction | KL divergence: +0.28 -> +0.27 | No clear effect |
| fix road-glance sensor noise lambda_R to best-guess SNR instead of inferring it (ablation) | approach 3 · Driver glance prediction | KL divergence | No clear effect |
Closest papers by meaning
- Learning Situational Driving (CVPR 2020 · similarity 0.87)
- Inverse Optimal Control Adapted to the Noise Characteristics of the Human Sensorimotor System (NeurIPS 2021 · similarity 0.86)
- Inverse decision-making using neural amortized Bayesian actors (ICLR 2025 · similarity 0.85)
- BAT: Behavior-Aware Human-Like Trajectory Prediction for Autonomous Driving (AAAI 2024 · similarity 0.85)
- DriveGPT: Scaling Autoregressive Behavior Models for Driving (ICML 2025 · similarity 0.84)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (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.