Monotonic Networks

By Joseph Sill. 1997

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

Learning Credible Models (2017, methodology): “…w ∗ | θ | [ i ] is the i th largest parameter sorted by magnitude; and ⊙ is the training time, Sun et al . explore using features identified as relevant Vapnik et al . explore the theory of learning with privileged information In the linear setting, the method simplifies to l 2 regularization on”
Monotonic Calibrated Interpolated Look-Up Tables (2015, background): “…kernel estimate post-process yes 2447 Ben-David (1995) tree penalize splits yes 125 Sill and Abu-Mostafa (1997) neural net penalize pairs no 550 Sill (1998) neural net constrain function yes 196 Kay and Ungar (2000) neural net constrain function yes 100 Potharst and Feelders (2002a) tree…”
Interpretable Machine Learning - A Brief History, State-of-the-Art and Challenges (2020, background): “Some approaches aim to make the parts of a model more interpretable with, for example, a monotonicity constraint [106] or a modified loss function for disentangling concepts learned by a convolutional neural network [130].”
Domain knowledge integration in data mining using decision tables: case studies in churn prediction (2009, background): “…to domain knowledge integration that is not explicitly incorporated by the three previous studies, is the monotonicity constraint (Ben-David, 1995; Sill, 1998; Feelders and Pardoel, 2003; Velikova and Daniels, 2004; Altendorf et al , 2005; Martens et al , 2006; Van Gestel et al , 2007; Martens…”
Rule learning with monotonicity constraints (2009, unknown): “(Gamarnik, 1998), medical diagnosis (Sill, 1998), survey data (Cao-Van & De Baets, 2004) and many others.”
Unconstrained Monotonic Neural Networks (2019, methodology): “ence. 1 Introduction Monotonic neural networks have been known as powerful tools to build monotone models of a response variable with respect to individual explanatory variables [Archer and Wang,1993,Sill, 1998,Daniels and Velikova,2010,Gupta et al.,2016,You et al.,2017]. Recently, strictly monotoni”
Rule Extraction from Support Vector Machines: An Overview of Issues and Application in Credit Scoring (2008, methodology): “…classiﬁcation techniques have been put forward to deal with monotonicity, such as for Bayesian learning [1], classiﬁcation trees [7,21,24], classiﬁcation rules [43] and neural networks [54,66]; e.g., in the medical diagnosis [47], house price prediction [65] and credit scoring [21, 54] domains.”
Monotonic Multi-layer Perceptron Networks as Universal Approximators (2005, background): “A monotonic function m(x), IR → IR [4] can be approximated by a sequence of adjacent, compact areas mk with mk(x) = const.”
Nearest Neighbour Classification with Monotonicity Constraints (2008, background): “Examples are: classification trees [7,12,22], neural networks [3,26], and Bayesian networks [1,13].”
Building Acceptable Classification Models (2010, background): “The application of such monotonicity constraints has been applied, among others, in the medical diagnosis [40], house price prediction [60], and credit scoring [16, 47] domains.”
Avoiding Resentment Via Monotonic Fairness (2019, methodology): “If we desire our function to be monotone nondecreasing with respect to every dimension of X, we can enforce this by ensuring all weights in the network are strictly positive, for example by applying some transformation τ : R → R+ [37].”
Solving Partially Monotone Problems with Neural Networks (2007, methodology): “In [11] it is shown in a case study on bond rating that the three-layer monotone networks perform better than a linear model and standard neural networks for problems where monotonicity is present in the domain.”