Hierarchical Invariance for Robust and Interpretable Vision Tasks at Larger Scales

arXiv:2402.15430v1 Announce Type: cross
Abstract: Developing robust and interpretable vision systems is a crucial step towards trustworthy artificial intelligence. In this regard, a promising paradigm considers embedding task-required invariant structures, e.g., geometric invariance, in the fundamental image representation. However, such invariant representations typically exhibit limited discriminability, limiting their applications in larger-scale trustworthy vision tasks. For this open problem, we conduct a systematic investigation of hierarchical invariance, exploring this topic from theoretical, practical, and application perspectives. At the theoretical level, we …

abstract applications artificial artificial intelligence arxiv cs.cv cs.lg embedding hierarchical image intelligence paradigm regard representation robust systems tasks trustworthy type vision