AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Risk-Driven Design of Perception Systems

Anthony Corso, Sydney Michelle Katz, Craig A Innes, Xin Du, Subramanian Ramamoorthy, Mykel Kochenderfer

NeurIPS 2022 · Probabilistic and scientific modeling › Dynamical systems · approach: Stable contractive dynamics · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
Add a risk-sensitive loss to perception trainingInverted-pendulum perception system · Inverted pendulummean time to failure: 234 ± 62 -> 492 ± 6Improved
Train on states sampled by risk-driven weightsInverted-pendulum perception system · Inverted pendulummean time to failure: 152 ± 89 -> 500Improved
Add risk-driven data generationYOLOv5s DAA perception system · Aircraft DAA encountersNMACs: 33%Improved
Combine risk-sensitive loss and risk-driven data generationYOLOv5s DAA perception system · Aircraft DAA encountersNMACs: 37%Improved
Add risk-sensitive lossYOLOv5s DAA perception system · Aircraft DAA encountersNMACsImproved
Set α=0.99 rather than α=0.00 for risk loss (ablation)Inverted-pendulum perception system · Inverted pendulummean time to failure: 492 ± 6 -> 391 ± 59Hurt
Set α=0.99 rather than α=0.20 for risk-driven data (ablation)Inverted-pendulum perception system · Inverted pendulummean time to failure: 500 ± 0 -> 273 ± 65Hurt
Apply risk-sensitive lossYOLOv5s DAA perception system · DAA validation setrecall: 0.38 ± 0.03 -> 0.37 ± 0.02No clear effect
Apply risk-driven data generationYOLOv5s DAA perception system · DAA validation setprecision: 0.78 ± 0.04 -> 0.77 ± 0.01No clear effect
Use risk-driven data generation and risk-sensitive lossYOLOv5s DAA perception system · DAA validation setrecall: 0.38 ± 0.03 -> 0.39 ± 0.01No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy