Learning from Data

By Doug Fisher, Hans-Joachim Lenz. 1996

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

On the Art and Science of Machine Learning Explanations (2018, methodology): “dations for each ∗H2O.ai, Mountain View, CA Figure 1: An augmented learning problem diagram in which several techniques create explanations for a credit scoring model. Adapted from Learning From Data [1]. method. Section 7 presents some general interpretability recommendations for practitioners. Sec”
Structured Pruning of Deep Convolutional Neural Networks (2015, neutral): “This results in comparable level of performance with fewer parameters and better generalization [4].”
D’ya Like DAGs? A Survey on Structure Learning and Causal Discovery (2021, background): “As such, the search problem is NP-hard [33], and this will later motivate the use of continuous optimization based algorithms for graph learning.”
Input feature selection for classification problems (2002, unknown): “The dataset is given as follows [24]:”
Is your dataset big enough? Sample size requirements when using artificial neural networks for discrete choice analysis (2018, unknown): “…testing set is an approximation of the so-called generalisation error, which is the key error for assessing an ANN's learning capability, because having an ANN with low generalisation error implies that the underlying data generating process has been well approximated (Abu-Mostafa et al., 2012).”
Machine Learning Control – Taming Nonlinear Dynamics and Turbulence (2016, background): “the propagation of the pressure from the body surface to the measurement device in tubes; (2) the computational time needed”
Machine Learning in Medicine (2015, background): “Machine learning is the scientific discipline that focuses on how computers learn from data.(1,2) It arises at the intersection of statistics, which seeks to learn relationships from data, and computer science, with its emphasis on efficient computing algorithms.”
Regularization via Mass Transportation (2017, background): “Thus, in spite of their many desirable theoretical properties, there is a consensus that “most of the (regularization) methods used successfully in practice are heuristic methods” [1].”
A Comprehensive Survey of Multiagent Reinforcement Learning (2008, unknown): “This feedback is less informative than in supervised learning, where the agent would be given the correct actions to take [9] (such information is, unfortunately, not always available).”
On closures for reduced order models—A spectrum of first-principle to machine-learned avenues (2021, unknown): “414 That being said, there are many methods for model identiﬁcation, starting with brute-force data interpolation.”
Beyond Accuracy, F-Score and ROC: A Family of Discriminant Measures for Performance Evaluation (2006, methodology): “As suggested by Bayes's theory (Cherkassky & Muller 1998; Duda, Hart, & Stork 2000), the measures listed in the survey section have the following effect: accuracy approximates how effective the algorithm is by showing the probability of the true value of the class label; in other words it assesses…”
DREAM: Diabetic Retinopathy Analysis Using Machine Learning (2014, methodology): “In this work, we selected the top 30 features shown in Table I out of 78 features that were ranked in decreasing order of feature weights generated by AdaBoost 12 [19].”