A neural network solves, explains, and generates university math problems by program synthesis and few-shot learning at human level

We demonstrate that a neural network pretrained on text and fine-tuned on code solves mathematics course problems, explains solutions, and generates questions at a human level. We automatically synthesize programs using few-shot learning and OpenAI’s Codex transformer and execute them to solve cours...

Descripción completa

Detalles Bibliográficos
Autores principales: Drori, Iddo, Zhang, Sarah, Shuttleworth, Reece, Tang, Leonard, Lu, Albert, Ke, Elizabeth, Liu, Kevin, Chen, Linda, Tran, Sunny, Cheng, Newman, Wang, Roman, Singh, Nikhil, Patti, Taylor L., Lynch, Jayson, Shporer, Avi, Verma, Nakul, Wu, Eugene, Strang, Gilbert
Formato: Online Artículo Texto
Lenguaje:English
Publicado: National Academy of Sciences 2022
Materias:
Physical Sciences
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9371704/
https://www.ncbi.nlm.nih.gov/pubmed/35917350
http://dx.doi.org/10.1073/pnas.2123433119

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9371704/
https://www.ncbi.nlm.nih.gov/pubmed/35917350
http://dx.doi.org/10.1073/pnas.2123433119

Ejemplares similares