PIRSA:23100032

Video URL

https://pirsa.org/23100032

Correlations, representations and transformers in physics and in AI

Kempf, A. (2023). Correlations, representations and transformers in physics and in AI. Perimeter Institute for Theoretical Physics. https://pirsa.org/23100032

Kempf, Achim. Correlations, representations and transformers in physics and in AI. Perimeter Institute for Theoretical Physics, Oct. 05, 2023, https://pirsa.org/23100032 

          @misc{ scivideos_PIRSA:23100032,
            doi = {10.48660/23100032},
            url = {https://pirsa.org/23100032},
            author = {Kempf, Achim},
            keywords = {Quantum Foundations},
            language = {en},
            title = {Correlations, representations and transformers in physics and in AI},
            publisher = {Perimeter Institute for Theoretical Physics},
            year = {2023},
            month = {oct},
            note = {PIRSA:23100032 see, \url{https://scivideos.org/index.php/pirsa/23100032}}
          }

Achim Kempf University of Waterloo

Talk numberPIRSA:23100032
DOI10.48660/23100032
Source RepositoryPIRSA
Collection
Talk Type Scientific Series
Subject

Abstract

I will begin by reviewing the general mathematical concept of representation. I will then show that representation theory is more generally applicable than one might expect, for example in quantum foundations, in quantum gravity and in machine learning. In those contexts, I will first talk about the notion of representation underlying phenomena of emergence in quantum gravity. I will then discuss how quantum reference frames might be viewed as representations. Finally, I will talk about how transformer models, such as GPT-4, construct representations of what they learn and, in that light, what it may take for machines to reach AGI or even consciousness.  

---

Zoom link: https://pitp.zoom.us/j/99349397588?pwd=T056VjZXRWZWT28zY3VOZHFDcmQ3QT09