Verifiably safe exploration for end-to-end reinforcement learning

N Hunt, N Fulton, S Magliacane, TN Hoang… - Proceedings of the 24th …, 2021 - dl.acm.org
Deploying deep reinforcement learning in safety-critical settings requires developing
algorithms that obey hard constraints during exploration. This paper contributes a first …

Reinbo: Machine learning pipeline search and configuration with bayesian optimization embedded reinforcement learning

X Sun, J Lin, B Bischl - arXiv preprint arXiv:1904.05381, 2019 - arxiv.org
Machine learning pipeline potentially consists of several stages of operations like data
preprocessing, feature engineering and machine learning model training. Each operation …

ReinBo: Machine learning pipeline conditional hierarchy search and configuration with Bayesian optimization embedded reinforcement learning

X Sun, J Lin, B Bischl - Machine Learning and Knowledge Discovery in …, 2020 - Springer
Abstract Machine learning pipeline potentially consists of several stages of operations like
data preprocessing, feature engineering and machine learning model training. Each …

[PDF][PDF] Predictive Analytics for Dynamic Pricing in Travel Bookings Using Machine Learning Pipelines

A Mantri - researchgate.net
Dynamic pricing, also referred to as surge pricing or time-based pricing, is a strategy where
businesses adjust product or service prices based on real-time market demand. This …

[PDF][PDF] Machine learning model selection with multi-objective Bayesian optimization and reinforcement learning: case studies on functional data analysis, pipeline …

X Sun - 2021 - d-nb.info
A machine learning system, including when used in reinforcement learning, is usually fed
with only limited data, while aimed at training a model with good predictive performance that …

Machine learning model selection with multi-objective Bayesian optimization and reinforcement learning

X Sun - 2021 - edoc.ub.uni-muenchen.de
A machine learning system, including when used in reinforcement learning, is usually fed
with only limited data, while aimed at training a model with good predictive performance that …