Cargando…

Socially situated artificial intelligence enables learning from human interaction

Regardless of how much data artificial intelligence agents have available, agents will inevitably encounter previously unseen situations in real-world deployments. Reacting to novel situations by acquiring new information from other people—socially situated learning—is a core faculty of human develo...

Descripción completa

Detalles Bibliográficos
Autores principales: Krishna, Ranjay, Lee, Donsuk, Fei-Fei, Li, Bernstein, Michael S.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: National Academy of Sciences 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9522339/
https://www.ncbi.nlm.nih.gov/pubmed/36122244
http://dx.doi.org/10.1073/pnas.2115730119
_version_ 1784800043411177472
author Krishna, Ranjay
Lee, Donsuk
Fei-Fei, Li
Bernstein, Michael S.
author_facet Krishna, Ranjay
Lee, Donsuk
Fei-Fei, Li
Bernstein, Michael S.
author_sort Krishna, Ranjay
collection PubMed
description Regardless of how much data artificial intelligence agents have available, agents will inevitably encounter previously unseen situations in real-world deployments. Reacting to novel situations by acquiring new information from other people—socially situated learning—is a core faculty of human development. Unfortunately, socially situated learning remains an open challenge for artificial intelligence agents because they must learn how to interact with people to seek out the information that they lack. In this article, we formalize the task of socially situated artificial intelligence—agents that seek out new information through social interactions with people—as a reinforcement learning problem where the agent learns to identify meaningful and informative questions via rewards observed through social interaction. We manifest our framework as an interactive agent that learns how to ask natural language questions about photos as it broadens its visual intelligence on a large photo-sharing social network. Unlike active-learning methods, which implicitly assume that humans are oracles willing to answer any question, our agent adapts its behavior based on observed norms of which questions people are or are not interested to answer. Through an 8-mo deployment where our agent interacted with 236,000 social media users, our agent improved its performance at recognizing new visual information by 112%. A controlled field experiment confirmed that our agent outperformed an active-learning baseline by 25.6%. This work advances opportunities for continuously improving artificial intelligence (AI) agents that better respect norms in open social environments.
format Online
Article
Text
id pubmed-9522339
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher National Academy of Sciences
record_format MEDLINE/PubMed
spelling pubmed-95223392022-09-30 Socially situated artificial intelligence enables learning from human interaction Krishna, Ranjay Lee, Donsuk Fei-Fei, Li Bernstein, Michael S. Proc Natl Acad Sci U S A Physical Sciences Regardless of how much data artificial intelligence agents have available, agents will inevitably encounter previously unseen situations in real-world deployments. Reacting to novel situations by acquiring new information from other people—socially situated learning—is a core faculty of human development. Unfortunately, socially situated learning remains an open challenge for artificial intelligence agents because they must learn how to interact with people to seek out the information that they lack. In this article, we formalize the task of socially situated artificial intelligence—agents that seek out new information through social interactions with people—as a reinforcement learning problem where the agent learns to identify meaningful and informative questions via rewards observed through social interaction. We manifest our framework as an interactive agent that learns how to ask natural language questions about photos as it broadens its visual intelligence on a large photo-sharing social network. Unlike active-learning methods, which implicitly assume that humans are oracles willing to answer any question, our agent adapts its behavior based on observed norms of which questions people are or are not interested to answer. Through an 8-mo deployment where our agent interacted with 236,000 social media users, our agent improved its performance at recognizing new visual information by 112%. A controlled field experiment confirmed that our agent outperformed an active-learning baseline by 25.6%. This work advances opportunities for continuously improving artificial intelligence (AI) agents that better respect norms in open social environments. National Academy of Sciences 2022-09-19 2022-09-27 /pmc/articles/PMC9522339/ /pubmed/36122244 http://dx.doi.org/10.1073/pnas.2115730119 Text en Copyright © 2022 the Author(s). Published by PNAS. https://creativecommons.org/licenses/by/4.0/This open access article is distributed under Creative Commons Attribution License 4.0 (CC BY) (https://creativecommons.org/licenses/by/4.0/) .
spellingShingle Physical Sciences
Krishna, Ranjay
Lee, Donsuk
Fei-Fei, Li
Bernstein, Michael S.
Socially situated artificial intelligence enables learning from human interaction
title Socially situated artificial intelligence enables learning from human interaction
title_full Socially situated artificial intelligence enables learning from human interaction
title_fullStr Socially situated artificial intelligence enables learning from human interaction
title_full_unstemmed Socially situated artificial intelligence enables learning from human interaction
title_short Socially situated artificial intelligence enables learning from human interaction
title_sort socially situated artificial intelligence enables learning from human interaction
topic Physical Sciences
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9522339/
https://www.ncbi.nlm.nih.gov/pubmed/36122244
http://dx.doi.org/10.1073/pnas.2115730119
work_keys_str_mv AT krishnaranjay sociallysituatedartificialintelligenceenableslearningfromhumaninteraction
AT leedonsuk sociallysituatedartificialintelligenceenableslearningfromhumaninteraction
AT feifeili sociallysituatedartificialintelligenceenableslearningfromhumaninteraction
AT bernsteinmichaels sociallysituatedartificialintelligenceenableslearningfromhumaninteraction