Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Tom Everitt
Renewable energy land rush needs fewer cowboys, more sheriffs
Consider the following scenario: An international, multi-billion dollar company proposes a new development and, as part of this project, it must obtain rights over a number of private properties in remote Australia. They casually allocate $80,000,000 for it and there is no Plan B – no land, no project.
[2412.04758] Measuring Goal-Directedness
arXiv:2412.04758 (cs) View PDF HTML (experimental) Comments: Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2412.04758 [cs.AI] (or arXiv:2412.04758v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2412.04758 Submission history From: James Fox [ view email] [v1] Fri, 6 Dec 2024 03:48:47 UTC (101 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Connected Papers Toggle Connected Papers (What is Connected Papers?)...
The Incentives that Shape Behaviour
Which variables does an agent have an incentive to control with its decision, and which variables does it have an incentive to respond to? We formalise these incentives, and demonstrate unique graphical criteria for detecting them in any single decision causal influence diagram. To this end, we introduce structural causal influence models, a hybrid of the influence diagram and structural causal model frameworks.
Designing agent incentives to avoid reward tampering
By Tom Everitt, Ramana Kumar, and Marcus Hutter From an AI safety perspective, having a clear design principle and a crisp characterization of what problem it solves means that we don’t have to guess which agents are safe. In this post and paper we describe how a design principle called current-RF optimization avoids the reward function tampering problem. Reinforcement learning (RL) agents are designed to maximize reward.
Understanding Agent Incentives with Causal Influence Diagrams
By Tom Everitt In our latest paper, we describe a new method for inferring agent incentives. The method is based on influence diagrams, which is a type of graphical model with special decision and utility nodes. In these, graphical criteria can be used to determine both agent observation incentives and agent intervention incentives. For humans, it is natural to think about intelligent systems as agents that strive to attain their goals by means of taking actions.
Scalable agent alignment via reward modeling: a research direction
Abstract: One obstacle to applying reinforcement learning algorithms to real-world problems is the lack of suitable reward functions. Designing such reward functions is difficult in part because the user only has an implicit understanding of the task objective. This gives rise to the agent alignment problem: how do we create agents that behave in accordance with the user's intentions?
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Tom
Contact Tom, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack