AI Paper Atlas

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

Temporally-Consistent Survival Analysis

Lucas Maystre, Daniel Russo

NeurIPS 2022 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory · 0 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
Replace landmarking regression targets with TCSR temporal-consistency targetsCox PH model · Synthetic treatment-effect Markov chain (Figure 1)Kendall-tau of estimated vs true treatment rankingImproved
Use TCSR instead of landmarking on PBC2 and RWCox PH model · PBC2, RWconcordance index and RMSEImproved
Use TCSR instead of landmarking to reach a given performance levelCox PH model · RWsample efficiency: 4×Improved
Use TCSR instead of landmarking at convergenceCox PH model · PBC2final model qualityImproved
Use landmarking and TCSR (full sequences) instead of the initial-state static baselineCox PH model · PBC2, RWconcordance index and RMSEImproved
Use landmarking instead of initial-state regressionCox PH model · AIDSconcordance indexHurt
Use TCSR instead of initial-state regressionCox PH model · AIDSconcordance index and RMSEMixed
Warm-start the inner optimizer with the previous iterate in TCSR (30 iterations)TCSR · All datasetsrunning time relative to landmarking: less than 30x landmarking timeImproved

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