AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Risk-Averse Active Sensing for Timely Outcome Prediction under Cost Pressure

Yuchao Qin, Mihaela van der Schaar, Changhee Lee

NeurIPS 2023 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization · 2 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
use risk-averse tail objective with adaptive intervals instead of constant-interval AS baselineRAS · synthetic DStimeliness d(delta=0.5): 0.522±0.012 -> 0.264±0.086Improved
use RAS instead of NLL adaptive sensingRAS · synthetic DStimeliness d(delta=0.5): 0.974±0.141 -> 0.264±0.086Improved
use RAS instead of ASACRAS · synthetic DSROC AUC: 0.605±0.096 -> 0.680±0.003Improved
use RAS instead of dense-sensing predictor FORAS · synthetic DStimeliness d(delta=0.5): 0.349±0.000 -> 0.264±0.086Improved
use RAS instead of dense sensing FORAS · synthetic DSacquisition cost: 31.000±0.000 -> 6.077±0.953Improved
focus on tail set (risk-averse) instead of risk-neutral ASRAS · synthetic DSacquisition cost: 4.501±0.497 -> 6.077±0.953Hurt
use RAS instead of AS baselineRAS · ADNIROC AUC: 0.704±0.023 -> 0.730±0.007Improved
use RAS instead of AS baselineRAS · ADNItimeliness d(delta=0.5): 2.357±0.375 -> 1.192±0.176Improved
use RAS instead of NLLRAS · ADNItimeliness d(delta=0.1): 1.040±0.149 -> 0.820±0.096Improved
use RAS instead of dense sensing FORAS · ADNItimeliness d(delta=0.1): 0.141±0.000 -> 0.820±0.096Hurt
use RAS instead of dense sensing FORAS · ADNIROC AUC: 0.747±0.000 -> 0.730±0.007Hurt
use RAS instead of dense sensing FORAS · ADNIacquisition cost: 26.865±0.000 -> 8.614±1.157Improved

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