What’s a Mindshake Time ?
Mindshake Time is designed as a discussion circle on a given research topic. It allows to discuss the latest advances, share thoughts, and initiate collaborations. It is intended for all experts who wish to broaden or confront their knowledge and vision, at the cutting edge of the state of the art.
Theme of this event :
To tackle interpretability in deep learning, we present a novel framework that jointly learns a predictive model and its associated interpretation model. The interpreter provides both local and global interpretability about the predictive model in terms of human-understandable high level attribute functions, with minimal loss of accuracy. This is achieved by a dedicated architecture and well chosen regularization penalties. A detailed pipeline to visualize the learnt features is also developed. Moreover, besides generating interpretable models by design, our approach can be specialized to provide post-hoc interpretations for a pre-trained neural network. We validate our approach against several state-of-the-art methods on multiple datasets and show its efficacy on both kinds of tasks.
Atelier réalisé le 10 Mai 2022
Paper’s link : https://arxiv.org/abs/2010.09345
Prerequisites : Basic knowledge in deep learning