Research map / Reinforcement learning / Visual and representation RL
Remotely Detectable Robot Policy Watermarking
Michael Amir, Manon Flageat, Amanda Prorok
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 |
|---|---|---|---|
| add GCC-PHAT offset handling to CoNoCo detection (ablation) | CoNoCo · Glimpse offset sweep (Onboard Sensors) | detection ROC AUC: AUC ≈ 1.0 | Improved |
| increase glimpse sequence length | CoNoCo · Glimpse sequence length sweep | detection ROC AUC: ROC AUC = 1 | Improved |
| move the Pendulum watermark band from [1.2,2.49] to [0.1,1.5] | CoNoCo · Inverted Pendulum | policy reward: B=[1.2,2.49] -> B=[0.1,1.5] | Improved |
| increase glimpse time jitter beyond the real-system value 0.002 | CoNoCo · Glimpse time-jitter sweep | detection ROC AUC: jitter 0.002 | No clear effect |
| rotate the camera plane of the glimpses by more than 60 degrees along x and y | CoNoCo · RoboMaster glimpse shift angle | detection ROC AUC: greater than 60° along both x and y | Hurt |
| add additive white noise attack with sigma_adv up to 2 | CoNoCo · HalfCheetah | detection AUC: sigma_adv = 2 | Mixed |
| add additive white noise attack with sigma_adv 0.25 | CoNoCo · Inverted Pendulum | detection AUC and reward: sigma_adv = 0.25 | Hurt |
Closest papers by meaning
- Prediction with Action: Visual Policy Learning via Joint Denoising Process (NeurIPS 2024 · similarity 0.82)
- Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning (NeurIPS 2024 · similarity 0.81)
- Learning to Act from Actionless Videos through Dense Correspondences (ICLR 2024 · similarity 0.81)
- Hidden in the Noise: Two-Stage Robust Watermarking for Images (ICLR 2025 · similarity 0.80)
- WMCopier: Forging Invisible Watermarks on Arbitrary Images (NeurIPS 2025 · similarity 0.80)
- Unbiased Watermark for Large Language Models (ICLR 2024 · similarity 0.80)
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.
