關於本電子書
2: Linear map: Introduces the basics of linear mapping and its role in reducing data dimensionality in machine learning.
3: Support vector machine: Learn how support vector machines apply dimensionality reduction in classification tasks and pattern recognition.
4: Principal component analysis: Delve into PCA's technique for transforming data into a set of linearly uncorrelated variables.
5: Isometry: Examine how isometric techniques preserve distances between points while reducing data dimensions.
6: Dimensionality reduction: Understand the broader scope of dimensionality reduction and its applications in various fields.
7: Semidefinite embedding: Study semidefinite programming and its connection to dimensionality reduction methods.
8: Kernel method: Discover the power of kernel methods in handling nonlinear relationships in data reduction.
9: Kernel principal component analysis: Explore KPCA’s capability to perform dimensionality reduction in a highdimensional feature space.
10: Numerical continuation: Learn how numerical continuation techniques assist in understanding highdimensional systems.
11: Spectral clustering: Understand how spectral clustering leverages dimensionality reduction to group similar data points.
12: Isomap: A look at Isomap, a technique that combines multidimensional scaling with geodesic distances for dimensionality reduction.
13: Johnson–Lindenstrauss lemma: Delve into the mathematics of the JohnsonLindenstrauss lemma, which ensures dimensionality reduction maintains geometric properties.
14: LinearnonlinearPoisson cascade model: Study how this model integrates linear and nonlinear methods in dimensionality reduction.
15: Manifold alignment: Learn about manifold alignment and its importance in aligning data from different domains in dimensionality reduction.
16: Diffusion map: Understand how diffusion maps use the diffusion process for dimensionality reduction in complex datasets.
17: Tdistributed stochastic neighbor embedding: Explore tSNE's ability to reduce dimensionality while preserving local structures in data.
18: Kernel embedding of distributions: Study how kernel embedding allows for dimensionality reduction on distributions, not just datasets.
19: Random projection: A practical approach to dimensionality reduction that relies on random projections for fast computation.
20: Manifold regularization: Learn about manifold regularization techniques and their impact on learning from highdimensional data.
21: Empirical dynamic modeling: Discover how empirical dynamic modeling aids in dimensionality reduction through time series data analysis.
