AISafety 2019                                          Keyword Index


Keyword Index


          Accessibility                                  91
          Adaptive design                                 8
          Adversarial Attacks                           112
          agent incentives                               22
          agent type                                     82
          AGI                                            44
          AGI safety                                     44
          AI for Safety                                   1
          AI Safety                                 75, 130
          AI safety                                     105
          AI Safety in Healthcare                       119
          AI Safety Landscape                             1
          AI Saftey                                       8
          AI value alignment                             75
          AIXI                                          105
          algorithmic fairness                           37
          artificial intelligence                        68
          Artificial Intelligence                        61
          Augmented Utilitarianism                       75
          Autonomous Driving                             61
          Autonomous Systems                             57
          autonomous vehicles                            82

          Bayesian learning                            105
          Behavioral Safety                             61
          Bias                                          91
          Black-Box Attacks                            112

          causal graphs                                105
          causality                                     44
          Clinical trial                                 8
          combinatorial testing using ontologies       123
          Commitments                                   98
          conservative                                  29
          Coordination and Cooperation                  98
          corrigibility                                 29
          counts as                                     68
          crowdsourcing                                 82

          deep learning                              37, 51
          Deep Neural Networks                           57
          Deep Q-learning from Demonstrations           112
          Deep Reinforcement Learning              112, 137
          Dependability                                  57

                                                                  1
AISafety 2019                                         Keyword Index


          DL                                           37
          Dynamic Safety Cases                          1

          embedded agency                             105
          Empathy                                     130
          Ethical Goal Function                        75
          ethics                                       82
          Explainable and Safe AI                     119

          frameworks                                   44

          Healthcare                                  119

          impact                                       29
          impact measures                              22
          influence diagrams                           44
          inverse reinforcement learning               37
          IRL                                          37

          learning                                     37
          Learning from Demonstrations                112

          Machine Ethics                              130
          machine learning                      15, 37, 51
          markov decision processes                     37
          Markov Decisions Processes                    98
          MDPs                                          37
          misalignment                                105
          ML                                            37
          modal logic                                   68
          moral dilemmas                                82
          multi-armed bandit                             8
          Multi-Task Learning                         137

          objective specification                      22

          Perception                                   57
          perturbations                                51
          Planning under Uncertainty                   98
          Policy Imitation                            112
          preferences                                  15

          Reinforcement Learning                   61, 130
          reinforcement learning           22, 29, 37, 105
          review                                        44
          RL                                            37

          Safety                                       57


2
AISafety 2019                                       Keyword Index


          Safety of Autonomy                           1
          Safety-Criticality                          61
          Sequential Triggers                       137
          side effects                            22, 29
          Software engineers                          91
          system testing for autonomous systems     123

          taxonomy                                  105
          test automation                           123
          Topological data analysis                  91
          transparency                               82
          trolley problem                            82
          trust                                      68

          Uncertainity                               57
          uncertainty                                37

          value alignment                            15
          value learning                             37
          verification                               68
          virtual reality                            82

          Watermarking                              137
          wireheading                               105




                                                               3