Research map / Time series / Causal and generative time series
Safe Active Learning with Future Viability Guarantees in Time-Series Models
Hyeonjun Park, Whiyoung Jung, Deunsol Yoon, Sunghoon Hong, Kanghoon Lee, Kyungjae Lee
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes safe active learning for time-series models that enforces future viability, not just one-step safety, so data collection avoids states with no safe continuation.
Problem
Safe active learning in time-series systems: one-step safety does not guarantee future viability, so an input that is immediately safe may lead the system into a state from which no safe continuation exists.
Approach
Extends safe active learning from one-step safety constraints to multi-step future viability guarantees, so that exploration inputs are chosen to keep the system in states with a safe continuation.
Closest papers by meaning
- Safely Learning Controlled Stochastic Dynamics (NeurIPS 2025 · similarity 0.88)
- Verified Safe Reinforcement Learning for Neural Network Dynamic Models (NeurIPS 2024 · similarity 0.86)
- A Physics-Informed Machine Learning Framework for Safe and Optimal Control of Autonomous Systems (ICML 2025 · similarity 0.85)
- Sampling-Based Safe Reinforcement Learning (NeurIPS 2026 · similarity 0.85)
- Learning Safe Control via On-the-Fly Bandit Exploration (ICML 2025 · similarity 0.84)
- Transductive Active Learning: Theory and Applications (NeurIPS 2024 · 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.
