The Evidence Framework Applied to Classification Networks

By David John Cameron MacKay. 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

Bayesian Neural Word Embedding (2016, neutral): “(16) and [15], respectively.”
Recasting Gradient-Based Meta-Learning as Hierarchical Bayes (2018, neutral): “The Laplace approximation (Laplace, 1986; MacKay, 1992a;b) is applicable in this case as it replaces a point estimate of an integral with the volume of a Gaussian centered at a mode of the integrand, thereby forming a local quadratic approximation.”
Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks (2020, methodology): “At the same time, in the case of Laplace approximations (MacKay, 1992b; Ritter et al., 2018), this treatment does not change the decision boundary of the trained network, so it has no negative effect on the predictive performance (cf. Figure 1).”
Bayesian Interpolation (1992, methodology): “Cross-validation, the orthodox "method of choice" (Eubank 19881, will be discussed more in Section 6.6 and MacKay (1992a).”
Customer base analysis: partial defection of behaviourally loyal clients in a non-contractual FMCG retail setting (2005, methodology): “Within the broad group of neural network architectures we select MacKay s Bayesian ARD neural network framework because it has the appealing property of providing a Bayesian hyperparameter per input variable, representing the importance of the variable (MacKay, 1992).”
Information Theory, Inference, and Learning Algorithms (2004, methodology): “A review paper on regression with complexity control using hierarchical Bayesian models is (MacKay, 1992a).”
Fast Marginal Likelihood Maximisation for Sparse Bayesian Models (2003, methodology): “We thus utilise the Laplace approximation procedure, as used in [6]:”
Preoperative prediction of malignancy of ovarian tumors using least squares support vector machines (2003, unknown): “On the other hand, a similar problem of finding good hyperparameters in the training of feedforward neural networks, has been tackled by applying the Bayesian framework [5,15,13].”