Nonlinear mapping
A nonlinear mapping is a function that does not preserve linear combinations of its inputs.
Math and Architectures of Deep Learning
A property of a function or layer whose output is not a linear transformation of its inputs.
- How it is used
- Uses nonlinear mappings to model curved decision boundaries and complex classification or estimation functions.
- How deep it goes
- Defines linearity mathematically, contrasts vector-length mapping, and connects sigmoid and layered architectures to expressive power.
- Examples
- Uses curved separators, sigmoid graphs, dog recognition, vector length, and quadratic regression implemented with PyTorch.
- What you can do after
- Apply it via PyTorch by adding nonlinear terms or nonlinear layers when linear models cannot fit the observed distribution.
Read how Math and Architectures of Deep Learning explains nonlinear mapping →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
A nonlinear transformation maps inputs to outputs in a way that cannot be represented as a linear function.
- How it is used
- Improves tabular prediction when feature effects or feature interactions do not follow straight-line relationships.
- How deep it goes
- Contrasts linear contributions with practical outlier evidence, then recommends nonlinear models or engineered features without deriving their internals.
- Examples
- Uses Tokyo Airbnb prices, bedroom counts, minimum-night values, prediction outliers, and per-feature contribution diagnostics.
- What you can do after
- Diagnose/debug with it by identifying linear-model failures and selecting feature transformations or nonlinear models.
Graph Neural Networks in Action
A parameterized operation that maps inputs to outputs without preserving linearity, allowing a neural network to model nonlinear relationships.
- How it is used
- Builds expressive node, edge, or graph embeddings within message-passing neural networks.
- How deep it goes
- Explains aggregation, learnable transformation, activation, and update steps, including ReLU and a general embedding formula.
- Examples
- Uses ReLU, multilayer perceptrons, neighbor aggregation, and employee recommendation or company-joining prediction.
- What you can do after
- Apply it through activation functions in message-passing layers to construct nonlinear graph embeddings and predictions.
Read how Graph Neural Networks in Action explains nonlinear mapping →
Deep Learning with PyTorch, Second Edition
A process whose behavior cannot be represented as a simple linear relation between inputs and outputs.
- How it is used
- Uses nonlinear mappings to let neural networks model complicated input/output relationships without specifying the exact function beforehand.
- How deep it goes
- Defines nonlinearity through activation functions, gradient-friendly behavior, saturation, and composed tanh examples, stopping short of dense theory.
- Examples
- Uses tanh neurons, summed bumps, two-layer compositions, Hardtanh approximations, and image-based dog detection.
- What you can do after
- Build it from scratch by composing linear units and differentiable activations to approximate complex input/output relationships.
Read how Deep Learning with PyTorch, Second Edition explains nonlinear mapping →
Statistics Slam Dunk
Nonlinearity is the property that a relationship cannot be represented adequately by a straight-line pattern or linear function.
- How it is used
- Uses Pareto charts to reveal uneven cause-and-effect patterns and support best-bang-for-the-buck corrective actions.
- How deep it goes
- Defines departure from linearity, relates it to diminishing returns, and demonstrates it visually without developing formal mathematical models.
- Examples
- Illustrates nonlinear distribution through NBA franchise games played and Pareto-chart output.
- What you can do after
- Evaluate nonlinear distributions by building and interpreting Pareto charts in R.
Statistics Every Programmer Needs
A nonlinear relationship is a relationship in which changes in one quantity are not represented by a linear function of another.
- How it is used
- Supports choosing transformations or nonlinear models when linear relationships inadequately represent forecasting or prediction problems.
- How deep it goes
- Explains transformations, outlier effects, and decision-tree behavior, but stops short of deriving nonlinear functions or algorithms.
- Examples
- Uses stock-price forecasting, logarithmic and square-root transformations, and a decision-tree comparison as concrete examples.
- What you can do after
- Evaluate whether transformations or decision trees better capture nonlinear patterns while checking assumptions, outliers, and overfitting.
Read how Statistics Every Programmer Needs explains nonlinear mapping →
Inside Deep Learning: Math, Algorithms, Models
A transformation inserted between linear steps to let a neural network represent functions that are not purely linear.
- How it is used
- Adds activation functions between linear operations to let networks represent more complex functions.
- How deep it goes
- Defines sigmoid and tanh, gives their ranges, and demonstrates tanh placement in a PyTorch network.
- Examples
- Uses sigmoid and tanh plots plus a two-layer nn.Sequential model with nn.Tanh.
- What you can do after
- Build it from scratch by placing nn.Tanh between PyTorch linear layers.
Read how Inside Deep Learning: Math, Algorithms, Models explains nonlinear mapping →
Related concepts
- activation functionproduced by · An activation function transforms a neural-network unit’s input into its output, typically introducing nonlinearity that increases the network’s representational capacity.
- 80-20 ruleequivalent to · The 80-20 rule is an informal heuristic that roughly 80 percent of an outcome may be associated with 20 percent of its contributing causes or effort.
- linear regressioncontrasts with · Linear regression is a statistical modeling method that estimates a continuous response as a linear combination of predictor variables and fitted coefficients.
- expressive powerproduces · Expressive power is the capacity of a model, language, or system to represent or convey a range and complexity of functions, patterns, distinctions, or ideas.
- mathematical functionkind of · A mathematical function is a mapping that assigns exactly one output to each permitted input in its domain.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.