Dictionary of Applied Machine Learning changelog

icon_url

— → https://dictionaryofml.org/favicon.svg

version

1.0.0 → 1.1.0

https://dictionaryofml.org/llms.txt

− > Open (CC BY 4.0) course edition of the Dictionary of Applied Machine Learning: 107 terms, each defined precisely and cross-referenced, with citations, a typeset PDF, and (for most terms) a Python demo that recomputes what the entry states. Maintained by Alexander Jung, Aalto University. DOI 10.5281/zenodo.21569296, ISBN 978-952-64-3013-3.− Full machine-readable content: [terms.json](https://dictionaryofml.org/terms.json) (one JSON file, every entry). Each term page carries a "Cite this entry" BibTeX block. Companion code (notebooks, LaTeX notation package, MCP server): https://github.com/alexjungaalto/dictionaryofml− Live MCP server (Model Context Protocol, streamable HTTP): https://dictionaryofml.org/mcp — read-only, no authentication. Four tools: list_all_terms, lookup_term, search_terms, get_related_terms. Add it as a custom connector, or `claude mcp add --transport http dictionaryofml https://dictionaryofml.org/mcp`. Documented at https://dictionaryofml.org/mcp.html− - [boosting](https://dictionaryofml.org/terms/boosting.html): Boosting is an iterative optimization method that sequentially combines base learners, referred to as weak learners.− - [bootstrap](https://dictionaryofml.org/terms/bootstrap.html): The bootstrap estimates how much the output of a machine learning method moves when its dataset is drawn again, using only the one dataset at hand.− - [confusion matrix](https://dictionaryofml.org/terms/cm.html): The confusion matrix of a hypothesis on a finite dataset with K label values is the K × K matrix whose entry in row y and column y counts the data points with true label y and prediction y.− - [covariance matrix](https://dictionaryofml.org/terms/covmtx.html): The covariance matrix of a random vector consists of the covariances between its entries.− - [decision tree](https://dictionaryofml.org/terms/decisiontree.html): A decision tree is a flowchart-like representation of a hypothesis map: a root node reads in the feature vector of a data point and evaluates a basic test on it.− - [feature importance](https://dictionaryofml.org/terms/featureimportance.html): An importance score assigns a number to each feature of a learned hypothesis.− - [independent and identically distributed assumption (i.i.d. assumption)](https://dictionaryofml.org/terms/iidasspt.html): The independent and identically distributed assumption (i.i.d.\ assumption) is the probabilistic model that represents the generation of a sequence of data points by independent and identically distributed RVs.− - [model](https://dictionaryofml.org/terms/model.html): In machine learning, the word \emph{model} is used in different ways.− - [robust principal component analysis (robust PCA)](https://dictionaryofml.org/terms/rpca.html): Robust principal component analysis (robust PCA) decomposes a data matrix into the sum of a matrix of low rank and a matrix with few nonzero entries.− - [sigmoid function](https://dictionaryofml.org/terms/sigmoid.html): The sigmoid function maps the real line into the interval between 0 and 1, turning the unbounded score that a classifier assigns to a data point into a probability.− - [supervised learning](https://dictionaryofml.org/terms/supervisedlearning.html): Supervised learning refers to machine learning methods that measure the quality of a hypothesis by comparing its prediction against a known label value.− - [singular value decomposition (SVD)](https://dictionaryofml.org/terms/svd.html): The singular value decomposition (SVD) is a factorization of a matrix A ∈ Rm × d of the form A = V Λ Uᵀ with orthogonal matrices V and U and a matrix Λ that is zero off its main diagonal, whose entries are nonnegative and referred to as sin− - [trusted execution environment (TEE)](https://dictionaryofml.org/terms/tee.html): A trusted execution environment (TEE) is a part of a computing system that runs code and holds data in isolation from the rest of the system.− - [test set](https://dictionaryofml.org/terms/testset.html): A test set consists of data points used to evaluate the performance of a hypothesis delivered by a machine learning method.− - [unsupervised learning](https://dictionaryofml.org/terms/unsupervisedlearning.html): Unsupervised learning is the machine learning setting in which the data points of a dataset carry features but no label, so nothing says what a correct answer for a data point would be.+ > Open (CC BY 4.0) course edition of the Dictionary of Applied Machine Learning: 114 terms, each defined precisely and cross-referenced, with citations, a typeset PDF, and (for most terms) a Python demo that recomputes what the entry states. Maintained by Alexander Jung, Aalto University. DOI 10.5281/zenodo.21569296, ISBN 978-952-64-3013-3.+ Full text in one request: [llms-full.txt](https://dictionaryofml.org/llms-full.txt) — every entry below, in full. Structured form: [terms.json](https://dictionaryofml.org/terms.json) (one JSON file, every entry). Single entry as Markdown: https://dictionaryofml.org/terms/<key>.md, where <key> is the last path segment of the term's URL. In all three, math is LaTeX with the book's macros expanded to standard LaTeX. Each term page carries a "Cite this entry" BibTeX block. Companion code (notebooks,+ Live MCP server (Model Context Protocol, streamable HTTP): https://dictionaryofml.org/mcp — read-only, no authentication. Six tools: list_all_terms, lookup_term, search_terms, get_related_terms, list_learning_paths (the author's curated reading orders per topic and audience), get_learning_path. Add it as a custom connector, or `claude mcp add --transport http dictionaryofml https://dictionaryofml.org/mcp`. Documented at https://dictionaryofml.org/mcp.html+ - [base learner](https://dictionaryofml.org/terms/baselearner.html): A base learner is the machine learning method that an ensemble method runs several times: an algorithm that reads a training set and returns a learned hypothesis.+ - [boosting](https://dictionaryofml.org/terms/boosting.html): Boosting is an ensemble method that sequentially combines simple base learners, also called weak learners.+ - [bootstrap](https://dictionaryofml.org/terms/bootstrap.html): The bootstrap is a stochastic algorithm that generates perturbed variants, referred to as bootstrap samples, of a given dataset.+ - [confusion matrix](https://dictionaryofml.org/terms/cm.html): The confusion matrix of a hypothesis on a finite dataset with K label values is the K × K matrix whose entry in row y and column ŷ counts the data points with true label y and prediction ŷ.+ - [correlation](https://dictionaryofml.org/terms/correlation.html): Correlation measures the strength and direction of the linear relationship between two real-valued RVs.+ - [covariance matrix](https://dictionaryofml.org/terms/covmtx.html): The covariance matrix of a random vector consists of the variances of the individual entries on its main diagonal and the covariances between different entries off the diagonal.+ - [decision boundary](https://dictionaryofml.org/terms/decisionboundary.html): The decision boundary of a classifier consists of the feature vectors that lie arbitrarily close to two or more decision regions.+ - [decision tree](https://dictionaryofml.org/terms/decisiontree.html): A decision tree is a flowchart-like representation of a hypothesis map: a root node receives the feature vector of a data point and evaluates a basic test on it.+ - [effective dimension](https://dictionaryofml.org/terms/effdim.html): The effective dimension of a hypothesis space is its size as seen through the probability distribution of the features: the number of model parameters that make a difference on the data points the distribution produces.+ - [fairness (machine learning)](https://dictionaryofml.org/terms/fairness.html): Fairness (machine learning) is the property of a learned hypothesis, or of the decision rule built on it, to treat data points with different values of a sensitive attribute, such as gender or age, in a comparable way.+ - [feature importance](https://dictionaryofml.org/terms/featureimportance.html): Feature importance assigns a number to each feature, an attribute of a data point, measuring how strongly a learned hypothesis relies on that feature when it maps a feature vector to a prediction.+ - [independent and identically distributed assumption (i.i.d. assumption)](https://dictionaryofml.org/terms/iidasspt.html): The independent and identically distributed assumption (i.i.d. assumption) is the probabilistic model that represents the generation of a sequence of data points by independent and identically distributed RVs.+ - [inverse](https://dictionaryofml.org/terms/inverse.html): An inverse of a function is a second function that undoes it.+ - [model](https://dictionaryofml.org/terms/model.html): In machine learning, the word model is used in different ways.+ - [robust principal component analysis (robust PCA)](https://dictionaryofml.org/terms/rpca.html): Robust principal component analysis (robust PCA) decomposes a feature matrix into a low-rank matrix, whose rows lie in a subspace of small dimension, and a sparse matrix that carries a few corrupted feature values.+ - [sigmoid function](https://dictionaryofml.org/terms/sigmoid.html): The sigmoid function maps the real line into the interval between 0 and 1.+ - [supervised learning](https://dictionaryofml.org/terms/supervisedlearning.html): Supervised learning is the machine learning setting that needs the label values of some data points.+ - [singular value decomposition (SVD)](https://dictionaryofml.org/terms/svd.html): The singular value decomposition (SVD) factors any matrix into an orthogonal matrix, a matrix that is zero off its main diagonal, and a second orthogonal matrix.+ - [trusted execution environment (TEE)](https://dictionaryofml.org/terms/tee.html): A trusted execution environment (TEE) is a part of a computing system that isolates code and data from other software.+ - [test set](https://dictionaryofml.org/terms/testset.html): A test set contains data points used to evaluate a hypothesis delivered by a machine learning method.+ - [trustworthy artificial intelligence (trustworthy AI)](https://dictionaryofml.org/terms/trustAI.html): Trustworthy artificial intelligence (trustworthy AI) names an artificial intelligence system that meets requirements beyond delivering accurate predictions.+ - [unsupervised learning](https://dictionaryofml.org/terms/unsupervisedlearning.html): Unsupervised learning refers to machine learning methods that do not require knowledge of the label of any data point.

• + Reading order: Wie funktioniert ChatGPT? (LLMs)
• + Reading order: KI-Grundlagen
• + Reading order: Erklärbare & vertrauenswürdige KI
• + Reading order: KI, Recht & AI Act
• + Reading order: KI in der Praxis (ML-Systeme)
• + Reading order: Mathematik der KI
• + accuracy
• + artificial neural network (ANN)
• + attention
• + autoencoder
• + backpropagation
• + bagging
• + base learner
• + baseline
• + binary classification
• + boosting
• + bootstrap
• + concept activation vector (CAV)
• + classification
• + clustering
• + confusion matrix
• + convex
• + correlation
• + covariance matrix
• + data
• + data augmentation
• + data imputation
• + data leakage
• + data point
• + dataset
• + decision boundary
• + decision tree
• + dimensionality reduction
• + distribution shift
• + effective dimension
• + eigenvalue
• + expectation–maximization (EM)
• + ensemble
• + empirical risk minimization (ERM)
• + eigenvalue decomposition (EVD)

Showing the first 40 of each.

• + list_learning_paths
• + get_learning_path