Reward reports for reinforcement learning
Building systems that are good for society in the face of complex societal effects requires a
dynamic approach. Recent approaches to machine learning (ML) documentation have …
dynamic approach. Recent approaches to machine learning (ML) documentation have …
The alignment ceiling: Objective mismatch in reinforcement learning from human feedback
N Lambert, R Calandra - arXiv preprint arXiv:2311.00168, 2023 - arxiv.org
Reinforcement learning from human feedback (RLHF) has emerged as a powerful technique
to make large language models (LLMs) more capable in complex settings. RLHF proceeds …
to make large language models (LLMs) more capable in complex settings. RLHF proceeds …
Entangled preferences: The history and risks of reinforcement learning and human feedback
Reinforcement learning from human feedback (RLHF) has emerged as a powerful technique
to make large language models (LLMs) easier to use and more effective. A core piece of the …
to make large language models (LLMs) easier to use and more effective. A core piece of the …
Designing Fiduciary Artificial Intelligence
S Benthall, D Shekman - Proceedings of the 3rd ACM Conference on …, 2023 - dl.acm.org
A fiduciary is a trusted agent that has the legal duty to act with loyalty and care towards a
principal that employs them. When fiduciary organizations interact with users through a …
principal that employs them. When fiduciary organizations interact with users through a …
Fleets on the streets: How number, affiliation and purpose of shared-lane automated vehicle convoys influence public perception and blame
Automated vehicles (AVs) may have broad uses in society, but some applications may be
more acceptable than others. Determining contexts in which AVs can acceptably operate is …
more acceptable than others. Determining contexts in which AVs can acceptably operate is …
Sociotechnical Specification for the Broader Impacts of Autonomous Vehicles
Autonomous Vehicles (AVs) will have a transformative impact on society. Beyond the local
safety and efficiency of individual vehicles, these effects will also change how people …
safety and efficiency of individual vehicles, these effects will also change how people …
[图书][B] Synergy of Prediction and Control in Model-based Reinforcement Learning
NO Lambert - 2022 - search.proquest.com
Abstract Model-based reinforcement learning (MBRL) has often been touted for its potential
to improve on the sample-efficiency, generalization, and safety of existing reinforcement …
to improve on the sample-efficiency, generalization, and safety of existing reinforcement …
Dynamic Documentation for AI Systems
S Mehta, A Rogers, TK Gilbert - arXiv preprint arXiv:2303.10854, 2023 - arxiv.org
AI documentation is a rapidly-growing channel for coordinating the design of AI technologies
with policies for transparency and accessibility. Calls to standardize and enact …
with policies for transparency and accessibility. Calls to standardize and enact …
Education for Sustainable Development and the Platform Society
M Shatkin - Digitalization, New Media, and Education for …, 2023 - igi-global.com
The chapter explores the risks of the platform society to sustainability and how education for
sustainability (ESD) can address these risks. The platform society is built on a business …
sustainability (ESD) can address these risks. The platform society is built on a business …