82 - Visual Reasoning, with Yoav Artzi
In this episode, Yoav Artzi joins us to talk about visual reasoning. We start by defining what visual reasoning is, then discuss the pros and cons of different tasks and datasets. We discuss some of the models used for visual reasoning and how they perform, before ending with open questions in this young, exciting research area.\n\nYoav Artzi: https://yoavartzi.com/\nNLVR: https://github.com/clic-lab/nlvr/tree/master/nlvr\nNLVR2: https://github.com/clic-lab/nlvr/tree/master/nlvr2\nCLEVR dataset: https://cs.stanford.edu/people/jcjohns/clevr/\nVQA: https://visualqa.org/\nGQA: https://cs.stanford.edu/people/dorarad/gqa/index.html\nNeural module networks: https://arxiv.org/abs/1511.02799