Combining active and semi-supervised learning for spoken language understanding

By Gökhan Tür, Dilek Z. Hakkani-Tür, Robert E. Schapire. 2005

271 citations recorded by the source corpus.

Subfields: Natural Language Processing, Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Speech & Audio

Citations in the corpus

Libri-Light: A Benchmark for ASR with Limited or No Supervision (2019, background): “In the semi-supervised setting, only a fraction of the dataset is labelled and the rest is unlabelled [1, 2], while in a distant supervision setting, the dataset is mostly or entirely unlabelled, but large quantities of”
From Theories to Queries (2011, methodology): “This simple approach is no more computationally expensive than performing inference, and has been shown to work well in a variety of applications (e.g., Lewis and Gale, 1994; Tong and Chang, 2001; Tür et al., 2005; Settles and Craven, 2008).”
Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds (2019, background): “[24] Gokhan Tur, Dilek Hakkani-Tür, and Robert E Schapire.”
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments (2016, methodology): “The effectiveness of self-training has been demonstrated in various areas, including spoken language understanding [31], handwritten digit and text classification [32], and sound event classification [33] .”
Semantic Parsing with Dual Learning (2019, background): “Beyond semantic parsing, the semisupervised and adaptive learnings are also typical in natural language understanding (Tur et al., 2005; Bapna et al., 2017; Zhu et al., 2014, 2018; Zhu and Yu, 2018).”
Self-Training: A Survey (2022, unknown): “Intuitively, thresholds should be set to high absolute values as pseudo-labeling examples with low confidence would increase chances of assigning wrong labels.”
Active Learning (2012, background): “, 2003), speech recognition (Tür et al., 2005), and cancer diagnosis (Liu, 2004), to name only a few.”
Active and Semisupervised Learning for the Classification of Remote Sensing Images (2014, methodology): “The techniques presented in [51] and [52] are special cases of this general combination strategy.”
Semi-Supervised Learning via Generalized Maximum Entropy (2010, methodology): “In machine learning, various methods to combine semi-supervised and active learning have been proposed to exploit the merits of both approaches (Zhu et al., 2003; Tür et al., 2005).”
Active Learning Literature Survey (2009, background): “…(Thompson et al., 1999; Settles and Craven, 2008), image classiﬁcation and retrieval (Tong and Chang, 2001; Zhang and Chen, 2002), video classiﬁcation and retrieval (Yan et al., 2003; Hauptmann et al., 2006), speech recognition (Tur et al., 2005), and cancer diagnosis (Liu, 2004) to name a few.”