Grasping Concept Spaces: Difference between revisions
From Computer Laboratory Group Design Projects
Jump to navigationJump to search
(Created page with "Client: Sam Henshall-West, JAID <sam.henshall-west@jaid.io> Today's most powerful machine learning models, including generative AI LLMs like ChatGPT, encode their knowledge as higher dimensional latent spaces. Vector-based concepts in that space can be clustered as regions on a 2D screen, for example using the t-SNE visualisation algorithm. Your task is to create a 3D version of t-SNE that is interactive, so users with VR headsets can literally “grasp” a concept, an...") |
No edit summary |
||
| Line 1: | Line 1: | ||
Client: Sam Henshall-West, JAID <sam.henshall-west@jaid.io> | Client: Sam Henshall-West, JAID <sam.henshall-west@jaid.io> | ||
Today's most powerful machine learning models, including generative AI LLMs like ChatGPT, encode their knowledge as higher dimensional latent spaces. Vector-based concepts in that space can be clustered as regions on a 2D screen, for example using the t-SNE visualisation algorithm. Your task is to create a 3D version of t-SNE that is interactive, so users with VR headsets can literally “grasp” a concept, and manipulate it to | Today's most powerful machine learning models, including generative AI LLMs like ChatGPT, encode their knowledge as higher dimensional latent spaces. Vector-based concepts in that space can be clustered as regions on a 2D screen, for example using the t-SNE visualisation algorithm. Your task is to create a 3D version of t-SNE that is interactive, so users with VR headsets can literally “grasp” a concept, and manipulate it to explore, fine-tune, or adjust the machine learning model. | ||
Latest revision as of 12:12, 29 October 2024
Client: Sam Henshall-West, JAID <sam.henshall-west@jaid.io>
Today's most powerful machine learning models, including generative AI LLMs like ChatGPT, encode their knowledge as higher dimensional latent spaces. Vector-based concepts in that space can be clustered as regions on a 2D screen, for example using the t-SNE visualisation algorithm. Your task is to create a 3D version of t-SNE that is interactive, so users with VR headsets can literally “grasp” a concept, and manipulate it to explore, fine-tune, or adjust the machine learning model.
