Everything about Generalization error
Measure of algorithm accuracy
For supervised learning applications in machine learning and statistical learning theory, generalization error is a measure of how accurately an algorithm is able to predict outcomes for previously unseen data. As learning algorithms are evaluated on finite samples, the evaluation of a learning algorithm may be sensitive to sampling error. As a result, measurements of prediction error on the current data may not provide much information about the algorithm's predictive ability on new, unseen data. The generalization error can be minimized by avoiding overfitting in the learning algorithm. The performance of machine learning algorithms is commonly visualized by learning curve plots that show estimates of the generalization error throughout the learning process. Wikipedia →
🧭 Start here: a 6-step reading path
- Read the overview: Generalization errorA 5-minute orientation
- Skim a recent paper: Estimates on the generalization error of physics-informed neural networks for approximating PDEsInfluential work since 2022
- Pick up a classic book: The Comedy of ErrorsThe most-reprinted book on the subject
- Read the landmark paper: Inference for the Generalization ErrorCited 1,021 times
- Go sideways: Statistical learning theoryA closely related topic
- Go sideways: Stability (learning theory)A closely related topic
📖 Definitions
No dictionary entry for this exact phrase.
🗂️ The subject at a glance
🏆 Landmark papers (most cited)
Inference for the Generalization Error
High-dimensional dynamics of generalization error in neural networks
An Efficient Method To Estimate Bagging's Generalization Error
Tightening Mutual Information-Based Bounds on Generalization Error
Generalization of Quantum Error Correction via the Heisenberg Picture
Input-dependent estimation of generalization error under covariate shift
Learning a new visuomotor transformation: error correction and generalization
κ-generalization of Gauss' law of error
🆕 Influential since 2022
Estimates on the generalization error of physics-informed neural networks for approximating PDEs
Generalisable 3D printing error detection and correction via multi-head neural networks
CoreDiff: Contextual Error-Modulated Generalized Diffusion Model for Low-Dose CT Denoising and Generalization
Generalization error of random feature and kernel methods: Hypercontractivity and kernel matrix concentration
Few-shot classification guided by generalization error bound
📚 Books worth reading
The Comedy of ErrorsWilliam Shakespeare · 1734
Ordeal by InnocenceAgatha Christie · 1958
AtonementIan McEwan · 2001
AirframeMichael Crichton · 1996
Dark hollowAnna Katharine Green · 1914
The InnocentHarlan Coben · 2004
The Innocent ManJohn Grisham · 2006
The ConfessionJohn Grisham · 2006
Ranked by number of editions on Open Library, a rough measure of how lasting a book is.
🔗 Related topics
- Statistical learning theory Statistical learning theory is a framework for machine learning drawing from the fields of statistics and functional analysis. Statistical learning theory…
- Stability (learning theory) Stability, also known as algorithmic stability, is a notion in computational learning theory of how a machine learning algorithm output is changed with…
- Learnable function class In statistical learning theory, a learnable function class is a set of functions for which an algorithm can be devised to asymptotically minimize the expected…
- Regularization (mathematics) In mathematics, statistics, finance, and computer science, particularly in machine learning and inverse problems, regularization is a process that converts…
- Supervised learning In machine learning, supervised learning (SL) is a type of machine learning paradigm where an algorithm learns to map input data to a specific output based…
- Early stopping In machine learning, early stopping is a form of regularization used to avoid overfitting when training a model with an iterative method, such as gradient…
- Gradient boosting Gradient boosting is a machine learning technique based on boosting in a functional space, where the target is pseudo-residuals instead of residuals as…
- Empirical risk minimization In statistical learning theory, the principle of empirical risk minimization defines a family of learning algorithms based on evaluating performance over…