On transfer learning using a MAC model variant

11/15/2018

∙

by   Vincent Marois, et al.

∙

0

∙

share

We introduce a variant of the MAC model (Hudson and Manning, CVPR 2018) with a simplified set of equations that achieves comparable accuracy, while training faster. We evaluate both models on CLEVR and CoGenT, and show that, transfer learning with fine-tuning results in a 15 point increase in accuracy, matching the state of the art. Finally, in contrast, we demonstrate that improper fine-tuning can actually reduce a model's accuracy as well.