B-cos Networks: Alignment is All We Need for Interpretability. (arXiv:2205.10268v1 [cs.CV])

We present a new direction for increasing the interpretability of deep neural
networks (DNNs) by promoting weight-input alignment during training. For this,
we propose to replace the linear transforms in DNNs by our B-cos transform. As
we show, a sequence (network) of such transforms induces a single linear
transform that faithfully summarises the full model computations. Moreover, the
B-cos transform introduces alignment pressure on the weights during
optimisation. As a result, those induced linear transforms become highly
interpretable and align …

alignment arxiv cv interpretability networks