Working for Esperdyne technologies, we were approached to help build an Expert Avatar(EA) system for use in museum and education settings. Using the baseline avatar of Albert Einstein, my tasks included creating sentiment models for use in the personality matrices of these historic teachers.
Deep Dive
The main challenge in this project lay in defining a personality itself. Should we create a personality system that changes with user interaction? Which primary documents should we include for semantic analysis? Is the psychological metric we’re using accurate?
50:50 Crystal | Rose
Latent Space Walk 50:50 Crystal | Rose
50:50 Crystal | Rose
75:25 Crystal | Rose
75:25 Crystal | Rose
Research Effective
With DataBrushing, we can take a closer look into how progressive growing GAN’s parse features/textures. This method let’s us peek a bit closer at the otherwise 'black box’ theory of generative networks. Although further research should be conducted, preliminary results from Databrushing also show support for the texture bias hypothesis of Geirhos 2018 et al. in generative models.
To read more, check back here June 2020 for the full medium article and Research Brief
