Information, Prediction, and Query by Committee

By Yoav Freund, H. Sebastian Seung, E. Shamir, Naftali Tishby. 1992

How researchers see this paper

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

Citations in the corpus

Deep Active Learning with a Neural Architecture Search (2018, background): “The literature on active learning in the context of classical models such as SVMs is extensive [3, 6, 27, 2, 1, 15], and clearly beyond the scope of this paper.”
A Sequential Algorithm for Training Text Classifiers (1994, background): “Freund, Seung, Shamir, and Tishby extend the QBC result to a wide range of classi(cid:12)er forms [13].”
Selective Sampling In Natural Language Learning (1995, methodology): “This approach was first advocated in theoretical work on the Query By Committee algorithm (Senng, Opper, and Sompolinsky, 1992; Freund et al., 1993), in which they assume a prior distribution on models and choose committee members randomly from the distribution restricted to the version space.”
Agnostic active learning (2006, methodology): “For example, the Query by Committee analysis ( Freund et al., 1993 ) assumes realizability (i.e., there exists a perfect classifier in a known set) and a correct Bayesian prior on the set of hypotheses.”
Improving Generalization with Active Learning (1994, background): “In these cases, oversampling of regions becomes a critical problem, and due to the inductive bias of the training algorithm, even a training-set size of only one may omit large regions of the domain.”
A sequential algorithm for training text classifiers (1994, background): “Freund, Seung, Shamir, and Tishby extend the QBC result to a wide range of classifier forms [13].”
Heterogeneous Uncertainty Sampling for Supervised Learning (1994, background): “Several algorithms have been proposed that base querying on filtering a stream of unlabeled instances rather than on creating artificial instances [6, 10, 20, 31].”
A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design (2017, background): “(Seung et al. 1992; Freund et al. 1993) for classification problems in the field of machine learning.”
Beyond neural scaling laws: beating power law scaling via data pruning (2022, background): “labeled [31, 32], rather than choose from a fixed set of inputs, as in data-pruning.”
Neural Network Ensembles, Cross Validation, and Active Learning (1994, background): “It is essentially a generalization of query by committee [6, 7] that was developed for classification problems.”