Skip to content

Latest commit

 

History

History
26 lines (25 loc) · 2.18 KB

SafeRL.md

File metadata and controls

26 lines (25 loc) · 2.18 KB

Reading List

Repository of reading lists on overview of AI Safety, Safe RL, and topics under the hood, along with selected paper summaries, and corresponding links

Risk, Transparency, Explainability

  • A Comprehensive Survey on Safe Reinforcement Learning [Paper] [Summary]
    • Javier Garcia, Fernando Fernandez, JMLR, 2015
  • Should Robots be Obedient? [Paper]
    • Smitha Milli, Dylan Hadfield-Menell, Anca Dragan, Stuart Russell
  • Enabling Robots to Communicate their Objectives [Paper]
    • Sandy H. Huang, David Held, Pieter Abbeel, Anca D. Dragan
  • Safe Model-based Reinforcement Learning with Stability Guarantees [Paper]
    • Felix Berkenkamp, Matteo Turchetta, Angela Schoellig, Andreas Krause, NIPS 2017
  • On ensuring that machines are well behaved [Paper] [Summary]
    • Philip S. Thomas, Bruno Castro da Silva, Andrew G. Barto, and Emma Brunskill
  • Safe Exploration in Markov Decision Processes [Paper]
    • Teodor Mihai Moldovan, Pieter Abbeel, ICML, 2012
  • The Off-Switch Game [Paper]
    • Dylan Hadfield-Menell, Anca Dragan, Pieter Abbeel, Stuart Russell
  • Concrete Problems in AI Safety [Paper]
    • Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, Dan Man´e
  • Constrained Policy Optimization [Paper]
    • Joshua Achiam, David Held, Aviv Tamar, Pieter Abbeel
  • Probabilistically Safe Policy Transfer [Paper]
    • David Held, Zoe McCarthy, Michael Zhang, Fred Shentu, Pieter Abbeel
  • Robust Covariate Shift Regression [Paper]
    • Xiangli Chen, Mathew Monfort, Anqi Liu, Brian D. Ziebart