Unsupervised Discrimination of Clustered Data via Optimization of Binary Information Gain

By N. Schraudolph, Terrence J. Sejnowski. 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 learning in neural networks: An overview (2014, neutral): “Many UL methods are designed to maximize information-theoretic objectives (e.g., Linsker, 1988; Barlow et al., 1989; MacKay and Miller, 1990; Plumbley, 1991; Schmidhuber, 1992b,c; Schraudolph and Sejnowski, 1993; Redlich, 1993; Zemel, 1993; Zemel and Hinton, 1994; Field, 1994; Hinton et al., 1995; D”
Discovering Gated Recurrent Neural Network Architectures (2018, methodology): “In this work, Binary Information Gain Optimization (BINGO, Schraudolph and Sejnowski (1993)) was used as an unsupervised objective.”
Neural Networks: Tricks of the Trade (2012, methodology): “the vector Ψ(t) will converge to the largest eigenvector of H and ‖Ψ(t)‖ to the corresponding eigenvalue [23, 14, 10].”
Alignment by Maximization of Mutual Information (1995, background): “BINGO was proposed to nd axes along which there is a bimodal distribution (Schraudolph and Sejnowski, 1993).”
Feature Extraction Through LOCOCODE (1999, unknown): “Some COCOFs explicitly favor near-factorial, minimally redundant codes of the input data (see, e.g., Watanabe, 1985; Barlow, Kaushal, & Mitchison, 1989; Linsker, 1988; Schmidhuber, 1992; Schmidhuber & Prelinger, 1993; Schraudolph & Sejnowski, 1993; Redlich, 1993; Deco & Parra, 1994).”
Unsupervised Learning in LSTM Recurrent Neural Networks (2001, methodology): “The BINGO algorithm clusters unlabeled data with linear adaptive discriminants stressing the gaps between clusters [ 48 ,49].”
Evolving Deep LSTM-based Memory Networks using an Information Maximization Objective (2016, methodology): “In this work, Binary Information Gain Optimization (BINGO, [21]) was used as an unsupervised objective.”