Joar Skalse - RankMe

24 papers A* 4A 1Journal 18Unranked 1

Year	Rank	Type	Title / Venue / Authors
2024	J	`jnl`	On the Limitations of Markovian Rewards to Express Multi-Objective, Risk-Sensitive, and Modal Tasks. CoRR Joar Skalse, Alessandro Abate
2024	J	`jnl`	Partial Identifiability and Misspecification in Inverse Reinforcement Learning. CoRR Joar Skalse, Alessandro Abate
2024	J	`jnl`	Partial Identifiability in Inverse Reinforcement Learning For Agents With Non-Exponential Discounting. CoRR Joar Skalse, Alessandro Abate
2024	J	`jnl`	Quantifying the Sensitivity of Inverse Reinforcement Learning to Misspecification. CoRR Joar Skalse, Alessandro Abate
2024	J	`jnl`	The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret. CoRR Lukas Fluri, Leon Lang, Alessandro Abate, Patrick Forré, David Krueger, Joar Skalse
2024	J	`jnl`	Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems. CoRR David Dalrymple, Joar Skalse, Yoshua Bengio, Stuart Russell, Max Tegmark, Sanjit A. Seshia, Steve Omohundro, Christian Szegedy, Ben Goldhaber, Nora Ammann, Alessandro Abate, Joe Halpern, Clark W. Barrett, Ding Zhao, Tan Zhi-Xuan, Jeannette M. Wing, Joshua B. Tenenbaum
2023	J	`jnl`	Goodhart's Law in Reinforcement Learning. CoRR Jacek Karwowski, Oliver Hayman, Xingjian Bai, Klaus Kiendlhofer, Charlie Griffin, Joar Skalse
2023	A*	`conf`	Misspecification in Inverse Reinforcement Learning. AAAI Joar Skalse, Alessandro Abate
2023	J	`jnl`	On The Expressivity of Objective-Specification Formalisms in Reinforcement Learning. CoRR Rohan Subramani, Marcus Williams, Max Heitmann, Halfdan Holm, Charlie Griffin, Joar Skalse
2023	A	`conf`	On the limitations of Markovian rewards to express multi-objective, risk-sensitive, and modal tasks. UAI Joar Skalse, Alessandro Abate
2023	J	`jnl`	STARC: A General Framework For Quantifying Differences Between Reward Functions. CoRR Joar Skalse, Lucy Farnik, Sumeet Ramesh Motwani, Erik Jenner, Adam Gleave, Alessandro Abate
2022	A*	`conf`	Defining and Characterizing Reward Gaming. NeurIPS Joar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger
2022	J	`jnl`	Defining and Characterizing Reward Hacking. CoRR Joar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger
2022	J	`jnl`	Invariance in Policy Optimisation and Partial Identifiability in Reward Learning. CoRR Joar Skalse, Matthew Farrugia-Roberts, Stuart Russell, Alessandro Abate, Adam Gleave
2022	A*	`conf`	Lexicographic Multi-Objective Reinforcement Learning. IJCAI Joar Skalse, Lewis Hammond, Charlie Griffin, Alessandro Abate
2022	J	`jnl`	Lexicographic Multi-Objective Reinforcement Learning. CoRR Joar Skalse, Lewis Hammond, Charlie Griffin, Alessandro Abate
2022	J	`jnl`	Misspecification in Inverse Reinforcement Learning. CoRR Joar Skalse, Alessandro Abate
2021	J	`jnl`	A General Counterexample to Any Decision Theory and Some Responses. CoRR Joar Skalse
2021	J	`jnl`	Is SGD a Bayesian sampler? Well, almost. J. Mach. Learn. Res. Chris Mingard, Guillermo Valle Pérez, Joar Skalse, Ard A. Louis
2021	A*	`conf`	Reinforcement Learning in Newcomblike Environments. NeurIPS James Bell, Linda Linsefors, Caspar Oesterheld, Joar Skalse
2021	—	`conf`	Safety Properties of Inductive Logic Programming. SafeAI@AAAI Gavin Leech, Nandi Schoots, Joar Skalse
2020	J	`jnl`	Is SGD a Bayesian sampler? Well, almost. CoRR Chris Mingard, Guillermo Valle Pérez, Joar Skalse, Ard A. Louis
2019	J	`jnl`	Neural networks are a priori biased towards Boolean functions with low entropy. CoRR Chris Mingard, Joar Skalse, Guillermo Valle Pérez, David Martínez-Rubio, Vladimir Mikulik, Ard A. Louis
2019	J	`jnl`	Risks from Learned Optimization in Advanced Machine Learning Systems. CoRR Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant