Dec 4, 2021 · In this work, we show the advantages of prototype representations for understanding and revising the latent space of neural concept learners.

Learning visual concepts from raw images without strong supervision is a challenging task. In this work, we show the advantages of prototype representations ...

Learning visual concepts from raw images without strong supervision is a challenging task. In this work, we show the advantages of prototype representations ...

Learning visual concepts from raw images without strong supervision is a challenging task. In this work, we show the advantages of prototype representations ...

In this work, we show the advantages of prototype representations for understanding and revising the latent space of neural concept learners.

Interactive Disentanglement: Learning Concepts by Interacting with their Prototype Representations. Supplementary Materials. 1 Hyperparameters and model ...

A novel framework for learning concept-grounded representations via weak super-vision and implicit prototype representations.

Discrete and semanticallygrounded representation is argued to be helpful for human understanding and abstract visual reasoning, enables fewshot or zero-shot ...

Interactive Disentanglement: Learning Concepts by Interacting with their Prototype Representations. Wolfgang Stammer Marius Memmel Patrick Schramowski ...

Interactive disentanglement: Learning concepts by interacting with their prototype representations. W Stammer, M Memmel, P Schramowski, K Kersting.