There is a long-standing interest in understanding the internal behavior of neural networks (Touret-zky and Pomerleau, 1989; Zhou et al., 2017; Raghu et al., 2017; Alishahi et al., 2017). Deep neural architectures for natural language processing (NLP) are often accompanied by explanations for their effectiveness, from general observations (e.g. RNNs can represent unbounded dependencies in a sequence) to specific arguments about linguistic phenomena (early layers encode lexical information, deeper layers syntactic). The recent ascendancy of DNNs is fueling efforts in the NLP community to explore these claims (Belinkov et al., 2017; Dalvi et al., 2017; Karpathy et al., 2015; Kadar et al., 2016; Kohn, 2015; Qian et al., 2016a). Previous work has tended to focus on easily-accessible representations like word or sentence embeddings (Kohn, 2015; Qian et al., 2016b; Adi et al., 2016), with deeper structure requiring more ad hoc methods to extract and examine (Belinkov and Glass, 2017; Poliaket al., 2018). In this work, we introduce Vivisect, a toolkit that aims at a general solution for broad and fine-grained monitoring in the major DNN frameworks, with minimal change to research patterns. Vivisect is general enough to serve as a less-polished version of the widely-used TensorBoard tool, but has several priorities that set it apart:? Minimal invasiveness (e.g. no SummaryOps) 1. Low resource use (only keep final metrics) 2. Uniform support for major DNN frameworks 3. Monitor performance on auxiliary tasks
展开▼
