Hubinger et al.'s “Risks from Learned Optimization in Advanced Machine Learning Systems”, one of our new core resources on the alignment problem, is now available on arXiv, the AI Alignment Forum, and LessWrong. In other news, we received an Ethereum donation worth $230,910 from Vitalik Buterin — the inventor and co-founder of Ethereum, and now our third-largest… Read more »
Posts By: Rob Bensinger
New paper: “Risks from learned optimization”
Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, and Scott Garrabrant have a new paper out: “Risks from learned optimization in advanced machine learning systems.” The paper’s abstract: We analyze the type of learned optimization that occurs when a learned model (such as a neural network) is itself an optimizer—a situation we refer to… Read more »
June 2019 Newsletter
Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, and Scott Garrabrant have released the first two (of five) posts on “mesa-optimization”: The goal of this sequence is to analyze the type of learned optimization that occurs when a learned model (such as a neural network) is itself an optimizer—a situation we refer to as… Read more »
May 2019 Newsletter
Updates A new paper from MIRI researcher Vanessa Kosoy, presented at the ICLR SafeML workshop this week: "Delegative Reinforcement Learning: Learning to Avoid Traps with a Little Help." New research posts: Learning "Known" Information When the Information is Not Actually Known; Defeating Goodhart and the "Closest Unblocked Strategy" Problem; Reinforcement Learning with Imperceptible Rewards The Long-Term Future Fund has announced twenty-three new… Read more »
New paper: “Delegative reinforcement learning”
MIRI Research Associate Vanessa Kosoy has written a new paper, “Delegative reinforcement learning: Learning to avoid traps with a little help.” Kosoy will be presenting the paper at the ICLR 2019 SafeML workshop in two weeks. The abstract reads: Most known regret bounds for reinforcement learning are either episodic or assume an environment without traps…. Read more »
April 2019 Newsletter
Updates New research posts: Simplified Preferences Needed, Simplified Preferences Sufficient; Smoothmin and Personal Identity; Example Population Ethics: Ordered Discounted Utility; A Theory of Human Values; A Concrete Proposal for Adversarial IDA MIRI has received a set of new grants from the Open Philanthropy Project and the Berkeley Existential Risk Initiative. News and links From the DeepMind safety team and Alex Turner: Designing Agent Incentives… Read more »
New grants from the Open Philanthropy Project and BERI
I’m happy to announce that MIRI has received two major new grants: A two-year grant totaling $2,112,500 from the Open Philanthropy Project. A $600,000 grant from the Berkeley Existential Risk Initiative. The Open Philanthropy Project’s grant was awarded as part of the first round of grants recommended by their new committee for effective altruism support:… Read more »
March 2019 Newsletter
Want to be in the reference class “people who solve the AI alignment problem”? We now have a guide on how to get started, based on our experience of what tends to make research groups successful. (Also on the AI Alignment Forum.) Other updates Demski and Garrabrant’s introduction to MIRI’s agent foundations research, “Embedded Agency,” is… Read more »
