By Matthew Rahtz, Vikrant Varma, Ramana Kumar, Zachary Kenton, Shane Legg, and Jan Leike. — By Matthew Rahtz, Vikrant Varma, Ramana Kumar, Zachary Kenton, Shane Legg, and Jan Leike. Tl;dr: ReQueST is an algorithm for learning objectives from human feedback on hypothetical behaviour. In this work, we scale ReQueST to complex 3D environments, and show that it works even with feedback sourced entirely from real…

By the Safety Analysis Team: Grégoire Delétang, Jordi Grau-Moya, Markus Kunesch, Tim Genewein, Rob Brekelmans, Shane Legg, and Pedro A… — By the Safety Analysis Team: Grégoire Delétang, Jordi Grau-Moya, Markus Kunesch, Tim Genewein, Rob Brekelmans, Shane Legg, and Pedro A. Ortega Read our paper here: https://arxiv.org/abs/2111.02907 We’re all familiar with risk-sensitive choices. As you look out of the window, you see a few gray clouds and no rain, but decide…

By Tom Everitt, Ryan Carey, Lewis Hammond, James Fox, Eric Langlois, and Shane Legg — By Tom Everitt, Ryan Carey, Lewis Hammond, James Fox, Eric Langlois, and Shane Legg Crossposted to the alignmentforum About 2 years ago, we released the first few papers on understanding agent incentives using causal influence diagrams. This blog post will summarize progress made since then. What are causal influence diagrams?

How can you tell if you have a good reward function? EPIC provides a fast and reliable way to evaluate reward functions. — By Adam Gleave, Michael Dennis, Shane Legg, Stuart Russell and Jan Leike. TL;DR: Equivalent-Policy Invariant Comparison (EPIC) provides a fast and reliable way to compute how similar a pair of reward functions are to one another. EPIC can be used to benchmark reward learning algorithms by comparing learned reward functions…

By Zachary Kenton, Tom Everitt, Laura Weidinger, Iason Gabriel, Vladimir Mikulik and Geoffrey Irving — By Zachary Kenton, Tom Everitt, Laura Weidinger, Iason Gabriel, Vladimir Mikulik and Geoffrey Irving Would your AI deceive you? This is a central question when considering the safety of AI, underlying many of the most pressing risks from current systems to future AGI. We have recently seen impressive advances in…

By the Safety Analysis Team: Grégoire Déletang, Jordi Grau-Moya, Miljan Martic, Tim Genewein, Tom McGrath, Vladimir Mikulik, Markus… — By the Safety Analysis Team: Grégoire Déletang, Jordi Grau-Moya, Miljan Martic, Tim Genewein, Tom McGrath, Vladimir Mikulik, Markus Kunesch, Shane Legg, and Pedro A. Ortega. TL;DR: To study agent behaviour we must use the tools of causal analysis rather than rely on observation alone. …

By Vladimir Mikulik , Grégoire Delétang, Tom McGrath, Tim Genewein, Markus Kunesch, Jordi Grau-Moya, Miljan Martic, Shane Legg, Pedro A… — By Grégoire Delétang, Tom McGrath, Tim Genewein, Vladimir Mikulik, Markus Kunesch, Jordi Grau-Moya, Miljan Martic, Shane Legg, Pedro A. Ortega TL;DR: In our recent paper we show that meta-trained recurrent neural networks implement Bayes-optimal algorithms. One of the most challenging problems in modern AI research is understanding the learned algorithms…

By Tom Everitt, Ramana Kumar, Jonathan Uesato, Victoria Krakovna, Richard Ngo, Shane Legg — By Tom Everitt, Ramana Kumar, Jonathan Uesato, Victoria Krakovna, Richard Ngo, Shane Legg In two new papers, we study tampering in simulation. The first paper describes a platform, called REALab, which makes tampering a natural part of the physics of the environment. …

Specification gaming is a behaviour that satisfies the literal specification of an objective without achieving the intended outcome. — Victoria Krakovna, Jonathan Uesato, Vladimir Mikulik, Matthew Rahtz, Tom Everitt, Ramana Kumar, Zac Kenton, Jan Leike, Shane Legg This article is cross-posted on the DeepMind website. Specification gaming is a behaviour that satisfies the literal specification of an objective without achieving the intended outcome. We have all had experiences with…