Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “Overfitting”

Overfitting

Candace Moore, Andrew Murphy, Dimitrios Toumpanakis · Radiopaedia.org · 2019

Cited by 1Open access

Overfitting

Cook, J, Ranstam, J · Oxford University Research Archive (ORA) (University of Oxford) · 2016

Overfitting is a key threat to the integrity and reliability of a statistical model.It is common to explore data looking for associations between potential risk factors and the development of a disease or outcome after treatment (e.g.influence of patient characteristics such as age and stage of disease upon recurrence of colorectal cancer after a partial coloctemy).Alternatively it may be desired to seek to build a prognostic model to enable prediction of future events (e.g.recurrence of disease or complications post-surgery).In such settings there are often a multiple of potential factors which could be modelled.However, the desire to include all available factors needs to be tempered.First, the dataset may not be large enough to support a model containing every factor.Second, some of the factors may be redundant and inclusion of them unnecessary and may even adversely impact (e.g.due to multicollinearity) upon the model by obscuring the true relationship between variables.Strategies are available (e.g.preselecting factors, centering, cross-validation) to minimise the possibility and impact of overfitting though ultimately replication of findings in another dataset is the only sure way to address the issue.Some more complex analyses (e.g.structural equation modelling) can be used which do not derive all of the information needed to support the analysis from the data; however this can only be done by greater prescription regarding relationships at the outset and the associated increased risk of having a self-fulfilling analysis.To paraphrase Antoine de Saint-Exupry, an analyst knows he has achieved perfection not when there is nothing left to add to the model, but when there is nothing left to take away.

Cited by 0Open access

Overfitting

Daniel Berrar, Werner Dubitzky · 2013

Cited by 0

Overfitting

Mutlu Yuksel, Yigit Aydede · 2025

Cited by 0

Quantifying Overfitting: Introducing the Overfitting Index

Sanad Aburass, Maha Abu Rumman · 2024

In the rapidly evolving domain of machine learning, ensuring model generalizability remains a quintessential challenge. Overfitting, where a model exhibits superior performance on training data but falters on unseen data, is a recurrent concern. This paper introduces the Overfitting Index (OI), a novel metric devised to quantitatively assess a model's tendency to overfit. Through extensive experiments on the Breast Ultrasound Images Dataset (BUS) and the MNIST dataset using architectures such as MobileNet, U-Net, ResNet, Darknet, and ViT-32, we illustrate the utility and discernment of the OI. Our results underscore the variable overfitting behaviors across architectures and highlight the mitigative impact of data augmentation, especially on smaller and more specialized datasets. The ViT-32's performance on MNIST further emphasizes the robustness of certain models and the dataset's comprehensive nature. By providing an objective lens to gauge overfitting, the OI offers a promising avenue to advance model optimization and ensure real-world efficacy.

Cited by 19

Quantifying Overfitting: Introducing the Overfitting Index

Sanad Aburass · arXiv (Cornell University) · 2023

In the rapidly evolving domain of machine learning, ensuring model generalizability remains a quintessential challenge. Overfitting, where a model exhibits superior performance on training data but falters on unseen data, is a recurrent concern. This paper introduces the Overfitting Index (OI), a novel metric devised to quantitatively assess a model's tendency to overfit. Through extensive experiments on the Breast Ultrasound Images Dataset (BUS) and the MNIST dataset using architectures such as MobileNet, U-Net, ResNet, Darknet, and ViT-32, we illustrate the utility and discernment of the OI. Our results underscore the variable overfitting behaviors across architectures and highlight the mitigative impact of data augmentation, especially on smaller and more specialized datasets. The ViT-32's performance on MNIST further emphasizes the robustness of certain models and the dataset's comprehensive nature. By providing an objective lens to gauge overfitting, the OI offers a promising avenue to advance model optimization and ensure real-world efficacy.

Cited by 7Open access

Moral Overfitting

Powers, Audrey · PhilPapers (PhilPapers Foundation)

This is a paper about model-building and overfitting in normative ethics. Overfitting is recognized as a methodological error in modeling in the philosophy of science and scientific practice, but this concern has not been brought to bear on the practice of normative ethics. I first argue that moral inquiry shares similarities with scientific inquiry in that both may productively rely on model-building, and, as such, overfitting worries should apply to both fields. I then offer a diagnosis of the problems of overfitting in moral inquiry and explain how our current practice seems worryingly susceptible to such problems. I conclude by giving suggestions for how we might avoid overfitting when doing normative ethics.

Cited by 0

Overfitting Explained

Paul R. Cohen, David Jensen, Producing C · 1996

Overfitting arises when model components are evaluated against the wrong reference distribution. Most modeling algorithms iteratively find the best of several components and then test whether this component is good enough to add to the model. We show that for independently distributed random variables, the reference distribution for any one variable underestimates the reference distribution for the the highest-valued variable; thus variate values will appear significant when they are not, and model components will be added when they should not be added. We relate this problem to the well-known statistical theory of multiple comparisons or simultaneous inference. 1 Iterative Modeling Algorithms Iterative modeling algorithms (IMAs) generate a search space M of models by repeatedly selecting a model m(\\Delta) 2 M and adding a component c i from a list of components C = c 1 ; c 2 ; :::; c n to m(\\Delta), producing m(\\Delta; c i ). For example, m(\\Delta) may be the regression equation y ...

Cited by 0

Moral Overfitting

Powers, Audrey · PhilPapers (PhilPapers Foundation)

This is a paper about model-building and overfitting in normative ethics. Overfitting is recognized as a methodological error in modeling in the philosophy of science and scientific practice, but this concern has not been brought to bear on the practice of normative ethics. I first argue that moral inquiry shares similarities with scientific inquiry in that both may productively rely on model-building, and, as such, overfitting worries should apply to both fields. I then offer a diagnosis of the problems of overfitting in moral inquiry and explain how our current practice seems worryingly susceptible to such problems. I conclude by giving suggestions for how we might avoid overfitting when doing normative ethics.

Cited by 0

Moral overfitting

Audrey Powers · Philosophical Studies · 2025

Abstract This is a paper about model-building and overfitting in normative ethics. Overfitting is recognized as a methodological error in modeling in the philosophy of science and scientific practice, but this concern has not been brought to bear on the practice of normative ethics. I first argue that moral inquiry shares similarities with scientific inquiry in that both may productively rely on model-building, and, as such, overfitting worries should apply to both fields. I then offer a diagnosis of the problems of overfitting in moral inquiry and explain how our current practice seems worryingly susceptible to such problems. I conclude by giving suggestions for how we might avoid overfitting when doing normative ethics.

Cited by 0Open access

Overfitting comparison.

Alina Bazarova (17138191), Marko Raseta (7384304) · Figshare · 2023

<p>The table reports average absolute difference between overfitting defined in “Results and discussion” for different methods.</p>

Cited by 0Open access

Index of Overfitting

Orhan, Mehemet, Orhan, Zeynep · 2024

This poster asserts that the purpose of the authors' research is to introduce an index for overfitting that makes use of suggestions to check for overfitting. It was presented at the UNT Research Day on October 4, 2024 in Denton, Texas.

Cited by 0

Illustration of overfitting.

Przemyslaw Baranski (3182034), Piotr Pietrzak (3182037) · Figshare · 2016

<p>Blue bullets—training data; red circles “unseen” (verification) data; continuous blue line—polynomial fitted accurately to 7 training points; red dashed lined—polynomial of optimal order fitted to the training data.</p>

Cited by 0Open access

Control for overfitting.

Jörn-Philipp Lies (531530), Ralf M. Häfner (531531), Matthias Bethge (146851) · Figshare · 2015

Cited by 0Open access

Overfitting evaluation results.

Haotong Han (20855036) · Figshare · 2025

<div> Objective An information security evaluation model based on the K-Means Clustering (KMC) + Decision Tree (DT) algorithm is constructed, aiming to assess its value in evaluating smart city (SC) security. Additionally, the impact of SCs on individuals’ mythical experiences is investigated. Methods An information security analysis model based on the combination of KMC and DT algorithms is established. A total of 38 SCs are selected as the research objects for practical analysis. The practical feasibility of the model is assessed using the receiver operating characteristic (ROC) curve, and its performance is compared with that of the Naive Bayes (NB), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), and Gradient Boosting Machine (GBM) classification methods. Lastly, a questionnaire survey is conducted to obtain and analyze individuals’ mythical experiences in SCs. Results (1) The area under the ROC curve is significantly higher than 0.9 (0.921 vs. 0.9). (2) Compared to the NB and LR algorithms, the security analysis model based on the combination of KMC and DT algorithms demonstrated higher true positive rate (TPR), accuracy, recall, F-Score, AUC-ROC, and AUC-PR. Additionally, the performance metrics of RF, SVM, and GBM are similar to those of the KMC+DT model. (3) When the attributes are the same, the difference in smart risk levels is small, while when the attributes are different, the difference in risk levels is significant. (4) The support rates for various types of new folk activities are as follows: offline shopping festivals (17.6%), New Year’s Eve celebrations (16.7%), Tibet tourism (15.6%), spiritual practices (16.2%), green leisure (16.0%), and suburban/rural tourism (15.8%). (5) High-risk cities (Grade A) showed stronger support for modern activities such as offline shopping festivals and green leisure, while low-risk cities (Grades C and D) tended to favor traditional cultural activities. Conclusion The algorithm model constructed in this work is capable of effectively evaluating the information security risks of SCs and has practical value. A good city image and mythological experience are driving the development of cities. </div>

Cited by 0Open access