liveBook · Manning

Machine learning

Also called ML

Machine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.

How 49 Manning books explain it

Optimization Algorithms: AI techniques for design, planning, and control problems

Machine learning is the computational study of methods that improve predictions or decisions by learning patterns from data or experience.

How it is used
Applies machine learning to produce solutions, configure algorithms, or guide combinatorial optimization searches.
How deep it goes
Defines learning through task, experience, and performance, then develops optimization applications and neural-network training; defers reinforcement learning.
Examples
Illustrates object recognition from vehicle images and penguin species classification using physical measurements.
What you can do after
Apply neural-network training via PySwarms to a penguin classification problem.

Read how Optimization Algorithms: AI techniques for design, planning, and control problems explains machine learning →

Build AI Drug Discovery Pipelines

Machine learning is a computational approach that learns patterns or predictive functions from data rather than relying solely on explicitly programmed rules.

How it is used
Applies machine learning to prioritize compounds and predict properties within drug-discovery and virtual-screening workflows.
How deep it goes
Explains training through loss, cost and parameter updates, and walks through model development while leaving production deployment and maintenance out of scope.
Examples
Uses cardiotoxicity prediction and hERG blocking to illustrate a ligand-based classification project.
What you can do after
Apply it via Scikit-Learn to train and evaluate a linear classifier, tune hyperparameters, and save the model for use on other data.

Read how Build AI Drug Discovery Pipelines explains machine learning →

Inside AI

Machine learning is a computational approach in which models learn patterns from data to perform tasks without being explicitly programmed for every case.

How it is used
Uses machine learning to examine predictive applications and the limits of data-driven decisions.
How deep it goes
Explains learned associations and validation through a clinical failure, without detailing model implementation.
Examples
Illustrates it with COVID-19 chest scans, pneumonia triage, and retail demand forecasting.
What you can do after
Evaluate predictions by considering data context, clinical consequences, and the need for expert validation.

Read how Inside AI explains machine learning →

Grokking Machine Learning

Machine learning is a field of artificial intelligence in which computers make decisions based on data rather than solely on explicitly specified rules.

How it is used
Uses data to help computers make decisions and predictions across tasks such as spam detection, diagnosis, and recommendations.
How deep it goes
Defines machine learning and outlines its decision-making framework and branches without detailing the algorithms.
Examples
Uses rain prediction, pet and email labels, house prices, spam detection, image recognition, recommendations, and diagnosis as examples.
What you can do after
Understand how data can support predictions and decisions, and distinguish learning types by their data and tasks.

Read how Grokking Machine Learning explains machine learning →

Natural Language Processing in Action, Second Edition

Machine learning is a computational approach in which algorithms derive patterns or behavior from data rather than relying solely on explicitly programmed rules.

How it is used
Uses machine learning to recognize text patterns and support chatbot and sentiment-analysis pipelines.
How deep it goes
Explains data-driven learning through comparisons with handcrafted rules, while deferring detailed model implementation.
Examples
Illustrates greeting recognition with regular expressions and sentiment scoring with labeled statements and Twitter hashtags.
What you can do after
Evaluate data-driven NLP against handcrafted rules while recognizing dataset quality and response reliability limitations.

Read how Natural Language Processing in Action, Second Edition explains machine learning →

Grokking Machine Learning, Second Edition

A computational technique in which systems learn patterns from data to make predictions or decisions without being explicitly programmed for every case.

How it is used
Applies data-based decision making to prediction, classification, grouping, recommendation, diagnosis, and other repeated data tasks.
How deep it goes
Defines data-driven decisions, models, algorithms, and the remember-formulate-predict process before later algorithms develop their own procedures.
Examples
Uses spam emails, house and stock prices, image recognition, recommendations, chatbots, self-driving cars, medical diagnosis, and market segmentation.
What you can do after
Evaluate/understand how past data becomes a model that predicts a new case through remember-formulate-predict.

Read how Grokking Machine Learning, Second Edition explains machine learning →

Deep Learning with R, Third Edition

Machine learning is an approach in which computer systems improve their performance on tasks by learning from data or experience rather than relying solely on explicitly programmed rules.

How it is used
Uses machine learning as the foundation for understanding deep learning and structuring projects from problem definition through deployment.
How deep it goes
Explains learning through inputs, expected outputs, feedback, and representations, then outlines problem framing, evaluation, and deployment.
Examples
Uses tagged vacation photos and speech transcripts to illustrate training, and fraud detection and photo search to illustrate project framing.
What you can do after
Evaluate and understand learning-based approaches, frame tasks around available data, and recognize the need for model evaluation.

Read how Deep Learning with R, Third Edition explains machine learning →

Demand Forecasting Best Practices

Machine learning is a computational approach in which models learn patterns from data to make predictions or decisions.

How it is used
Generates demand-forecast baselines from historical demand and demand drivers, reducing manual planning work.
How deep it goes
Defines supervised demand forecasting, contrasts statistical models, and details evaluation, interpretability, and launch pitfalls without model construction.
Examples
Uses seasonal versus non-seasonal products, weather and promotions, and a 2018–2022 training versus 2023 test split.
What you can do after
Evaluate/understand machine-learning demand forecasts by selecting objectives, testing fairly, benchmarking results, and interpreting scenarios cautiously.

Read how Demand Forecasting Best Practices explains machine learning →

The Art of AI Product Development

Machine learning is a field of artificial intelligence in which systems learn patterns from data to make predictions or decisions.

How it is used
Supports data-driven product decisions through prediction, classification, and personalization.
How deep it goes
Explains numerical preprocessing and contrasts learned models with rules, without detailing model-training mechanics.
Examples
Uses fashion recommendations, one-hot word encoding, and demand forecasting to illustrate applications and inputs.
What you can do after
Evaluate whether learned models or rules fit a product problem, and communicate forecast uncertainty to customers.

Read how The Art of AI Product Development explains machine learning →

Advanced Algorithms and Data Structures

Machine learning is a computational technique in which mathematical models infer characteristics or patterns of a system from data.

How it is used
Uses machine learning to frame clustering as a way to discover groups in unlabeled customer data.
How deep it goes
Defines learned models and training-feedback types, then connects them to clustering and gradient descent without deriving a general training procedure.
Examples
Illustrates learning with customer segmentation, object detection, price estimation, chess, robotics, and clustering algorithms.
What you can do after
Apply it via k-means or DBSCAN to group an unlabeled dataset, while choosing between implementation and an available library.

Read how Advanced Algorithms and Data Structures explains machine learning →

Delivering Machine Learning Projects: From design to deployment

Machine learning is the computational approach of learning models from data through algorithms rather than specifying all model rules manually.

How it is used
Frames machine learning as a way to deliver models that address team-scale business problems.
How deep it goes
Defines learning and inference, compares machine and human model building, and sketches delivery without detailing individual algorithms.
Examples
Uses a bike-store sales and inventory project and a dog-versus-cat model to illustrate business decisions and approximate inference.
What you can do after
Evaluate how data, user needs, model handling, and ethical risks affect an ML project.

Read how Delivering Machine Learning Projects: From design to deployment explains machine learning →

Fight Fraud with Machine Learning

Machine learning is the computational approach of learning predictive or decision-making behavior from data rather than specifying all behavior explicitly.

How it is used
Supports scalable fraud decisions by learning from past fraud data when manually specified rules cannot keep up.
How deep it goes
Introduces labeled feature-to-output learning, binary classification, feature engineering, and deep-learning contrast without detailing individual algorithms.
Examples
Uses dog-versus-cat images, phishing emails, spam filtering, and online transactions to illustrate learned decisions.
What you can do after
Evaluate/understand when ML or rules suit fraud detection, including trade-offs in scale, interpretability, and latency.

Read how Fight Fraud with Machine Learning explains machine learning →

Data-Driven Cybersecurity

A machine learning algorithm enables a system to learn patterns from data and make predictions or decisions without explicitly programmed rules.

How it is used
Supports proactive threat detection and risk forecasting by finding patterns and anomalies in security data.
How deep it goes
Names model families and security applications, while deferring a deep dive into statistics and AI.
Examples
Uses a financial institution’s before-and-after detection dashboard and a healthcare executive presentation with response-time and investment figures.
What you can do after
Evaluate/understand it by checking detection gains against business goals, data quality, and model reliability.

Read how Data-Driven Cybersecurity explains machine learning →

Investing for Programmers

Machine learning is a field of computing in which algorithms learn patterns from data to make predictions or decisions.

How it is used
Deploys machine learning to extract financial insights and support investment decisions, especially within narrowly defined analytical use cases.
How deep it goes
Defines supervised and unsupervised learning and discusses market limitations, but stops short of detailed algorithm internals.
Examples
Demonstrates labeled stock-direction prediction and one-year NVIDIA forecasts generated from yfinance data.
What you can do after
Evaluate/understand when financial machine learning can provide narrow insights and when market volatility makes forecasts risky.

Read how Investing for Programmers explains machine learning →

Hidden Influences

A computational approach in which models use data and interactions to learn patterns for generating predictions or recommendations.

How it is used
Uses machine learning to explain how recommender systems learn from interactions and how models can help classify content.
How deep it goes
Explains learning through examples, preference updates, and layered representations, while noting that embeddings and individual recommendations can be hard to interpret.
Examples
Illustrates learning with spam filtering, engagement with followed accounts, and classification of policy-related posts.
What you can do after
Evaluate and understand how learned recommendations reflect observed interactions, can miss interests, and may be difficult to explain.

Read how Hidden Influences explains machine learning →

First-Party Data Activation

Machine learning is a computational approach in which models learn patterns from data to make predictions or decisions.

How it is used
Supports predictive marketing decisions and personalization using patterns in customer data.
How deep it goes
Explains marketing applications and architectural placement at a decision-making level, without model-building details.
Examples
Illustrates retail retention through purchase frequency, email engagement, adjusted send times, and personalized subject lines.
What you can do after
Evaluate how machine learning predictions can inform retention, recommendations, and campaign timing.

Read how First-Party Data Activation explains machine learning →

Acing the CCNA Exam: Volume 2 Advanced Networking and Security

Machine learning is a computational technique in which systems learn patterns from data to make predictions or decisions without being explicitly programmed for each case.

How it is used
Applies data-driven learning to network automation, performance improvement, security analysis, and operational decision-making.
How deep it goes
Defines ML, distinguishes learning types, and connects applications to networking without detailing algorithms, configuration, or implementation internals.
Examples
Uses labeled cat-and-dog images, clustering, traffic forecasting, predictive maintenance, capacity planning, and security threat prediction.
What you can do after
Evaluate/understand how learning types and ML applications support predictive analysis and automation in network operations.

Read how Acing the CCNA Exam: Volume 2 Advanced Networking and Security explains machine learning →

Math and Architectures of Deep Learning

A paradigm for learning a parameterized mapping from inputs to outputs by training on data rather than by writing explicit step-by-step rules.

How it is used
Frames classification and estimation as learned input-to-output transformations rather than explicit programs.
How deep it goes
Develops function approximation, parameterized architectures, training, inference, geometry, and expressive power, while deferring vectors and matrices.
Examples
Uses a cat brain with hardness and sharpness to produce threat scores and three behavioral decisions.
What you can do after
Evaluate/understand how training data determines model weights for later inference on classification or estimation tasks.

Read how Math and Architectures of Deep Learning explains machine learning →

Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

A field of computing in which algorithms learn patterns from data to make predictions or decisions.

How it is used
Applies classical predictive methods to business-oriented tabular-data problems and compares them with deep learning.
How deep it goes
Explains feature engineering and evaluates simplicity, transparency, and efficacy, without deriving model mathematics.
Examples
Uses New York City Airbnb price classification and a local real-estate time-on-market prediction scenario.
What you can do after
Evaluate/understand classical machine learning versus deep learning using simplicity, transparency, and efficacy for tabular problems.

Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains machine learning →

Data Without Labels

An algorithm that enables a computer system to learn patterns or behavior from data or interaction rather than being explicitly programmed for every outcome.

How it is used
Uses historical data to find patterns and support decisions about defined business problems.
How deep it goes
Defines the technique and distinguishes four learning approaches, while deferring algorithm internals and mathematical foundations to further treatment.
Examples
Illustrates recommendations with online shopping and travel offers, and contrasts learning approaches through prediction, segmentation, and autonomous navigation.
What you can do after
Apply it via Python to investigate a defined business problem, prepare data, and select an appropriate learning approach.

Read how Data Without Labels explains machine learning →

Deep Learning with Python, Third Edition

A class of methods that learn to map inputs to outputs by searching for useful representations and rules from example data and feedback.

How it is used
Introduces learning from examples as an alternative to handcoding rules for tasks such as image tagging.
How deep it goes
Explains training feedback and data representations conceptually, without specifying a training algorithm.
Examples
Uses human-tagged vacation photos and RGB versus HSV image representations to illustrate learning and task-specific encoding.
What you can do after
Understand how examples and feedback guide a model toward task-specific outputs.

Read how Deep Learning with Python, Third Edition explains machine learning →

AI-Powered Search

Machine learning is a subset of artificial intelligence that trains models on data to perform tasks based on patterns learned from that data.

How it is used
Uses machine learning to automate search configuration and improve relevance across the search pipeline.
How deep it goes
Defines machine learning through learning from data and outlines search applications, while deferring automated ranking implementation.
Examples
Uses manually tuned synonym lists and field weights to illustrate search configuration that machine learning can help automate.
What you can do after
Evaluate and understand where machine learning can automate relevance features and other search tasks.

Read how AI-Powered Search explains machine learning →

Outlier Detection in Python

Machine learning is the computational approach of learning patterns or predictive relationships from data rather than specifying all rules explicitly.

How it is used
Frames data-driven prediction and pattern discovery, especially unsupervised outlier detection across tabular and time-series records.
How deep it goes
Defines supervised, unsupervised, and deep learning, then connects methods to algorithms and practical detection workflows without detailed theory.
Examples
Uses labeled animal images, fraud records, sensor readings, network activity, and astronomical observations as motivating examples.
What you can do after
Apply machine learning via Python, NumPy, pandas, scikit-learn, and PyOD to develop and evaluate detection solutions.

Read how Outlier Detection in Python explains machine learning →

The AI Pocket Book

A field that builds systems that learn patterns from data to perform tasks.

How it is used
Creates LLMs and other AI systems by learning model parameters from data rather than specifying every behavior manually.
How deep it goes
Explains architecture, parameters, training, learning types, and explainability, but stops short of training algorithms or mathematical derivations.
Examples
Uses LLM next-token prediction, embedding construction, PyTorch layers, and product-purchase or machine-replacement predictions as examples.
What you can do after
Evaluate/understand how data, architectures, parameters, and training produce models, while recognizing when explainability limits their use.

Read how The AI Pocket Book explains machine learning →

How Large Language Models Work

Machine learning is the field of building computational systems that improve their performance on tasks by deriving patterns or rules from data rather than receiving all rules explicitly.

How it is used
Frames machine learning as the foundation for understanding how large language models derive capabilities from data.
How deep it goes
Explains learning through examples, algorithm design, scaling, embeddings, and trade-offs, but defers detailed mathematical and operational mechanics.
Examples
Uses cat-image classification, convolutional and recurrent networks, XGBoost, language embeddings, similarity plots, clustering, and outlier detection.
What you can do after
Apply it via embeddings and algorithms that group similar texts or identify unusual ones.

Read how How Large Language Models Work explains machine learning →

Grokking AI Algorithms, Second Edition

Machine learning is the field of building systems that learn patterns or behavior from data rather than relying solely on explicitly programmed rules.

How it is used
Uses data patterns to support predictions, decisions, and understanding of relationships.
How deep it goes
Defines three learning approaches and covers data preparation and implementations of linear regression and decision-tree learning.
Examples
Illustrates pattern-based rent estimation with apartment features and contrasts labeled fruit examples with grouping similar fruits.
What you can do after
Builds linear regression and decision-tree learning algorithms for prediction and classification from scratch.

Read how Grokking AI Algorithms, Second Edition explains machine learning →

Starting Data Analytics with Generative AI and Python

Machine learning is the computational use of algorithms that learn patterns from data to make predictions, classifications, or decisions.

How it is used
Turns structured language data into learned patterns or predictions for sentiment analysis and other text-mining tasks.
How deep it goes
Explains practical model use while deferring deep mechanics such as matrix multiplication.
Examples
Uses sentiment analysis across social media, websites, emails, company documents, and academic publications as application contexts.
What you can do after
Apply machine learning to sentiment analysis through pretrained models, libraries, pipelines, and the Hugging Face platform.

Read how Starting Data Analytics with Generative AI and Python explains machine learning →

Machine Learning Algorithms in Depth

An algorithm that learns rules or predictive relationships from data rather than receiving all task instructions explicitly.

How it is used
Frames learning rules from data as an alternative to supplying every task instruction explicitly.
How deep it goes
Defines a supervised mapping and derives average misclassification loss, but does not specify an optimizer or implement a complete model here.
Examples
Contrasts learned rules with sorting instructions and uses stock-price prediction, data clusters, and a training–testing diagram.
What you can do after
Evaluate/understand it to select algorithms, explain results, troubleshoot models, and adapt algorithms to new situations.

Read how Machine Learning Algorithms in Depth explains machine learning →

Grokking AI Applications

An area of artificial intelligence in which machines learn patterns or behavior autonomously from data rather than following only fixed rules.

How it is used
Frames machine learning as the mechanism that lets AI learn autonomously from data and supports generative AI.
How deep it goes
Defines three learning types and staged language-model training, but stops short of detailed algorithm internals.
Examples
Uses customer-purchase prediction, toy-shape grouping, child color learning, walking by trial and error, and ChatGPT training.
What you can do after
Evaluate/understand how data-driven prediction differs from fixed rules and how ChatGPT combines learning stages.

Read how Grokking AI Applications explains machine learning →

Vibe Engineering

Machine learning is the field and set of computational methods in which models learn patterns from data to make predictions or decisions.

How it is used
Uses machine learning to frame business prediction problems within LLM-assisted data science.
How deep it goes
Introduces dataset selection and prediction inputs while deferring detailed machine learning and deep learning explanations.
Examples
Compares real and synthetic medicine datasets and specifies drug-price prediction features.
What you can do after
Evaluate/understand dataset suitability and the inputs to a drug-price prediction problem.

Read how Vibe Engineering explains machine learning →

Transfer Learning for Natural Language Processing

A field of computing in which algorithms learn patterns or transformations from data rather than being explicitly programmed for every scenario.

How it is used
Positions learning from data as an alternative to hand-coded rules and connects it to natural language processing.
How deep it goes
Explains learning through input-output pairs, loss functions, hypothesis sets, and decision boundaries without a worked implementation.
Examples
Uses positive and negative sentence labels and an arc-shaped decision boundary to illustrate classification.
What you can do after
Understand how training data, a loss function, and a hypothesis set guide learning toward generalization.

Read how Transfer Learning for Natural Language Processing explains machine learning →

Machine Learning Platform Engineering

Machine learning is the computational discipline of learning patterns from data to make predictions or decisions.

How it is used
Deploys learned patterns to solve business problems through prediction or decision-making systems.
How deep it goes
Explains the production life cycle and experimentation loops, but stops short of detailed algorithms, backpropagation, or mathematical derivations.
Examples
Uses OCR, iterative model evaluation, orchestrated pipelines, and frameworks such as TensorFlow, PyTorch, and scikit-learn.
What you can do after
Build reliable ML systems by selecting suitable problems, iterating on data and models, and automating reproducible workflows.

Read how Machine Learning Platform Engineering explains machine learning →

Designing Deep Learning Systems: A guide for software engineers

Machine learning is the field of methods that learn patterns or decision functions from data rather than relying solely on explicitly programmed rules.

How it is used
Positions machine learning as both a decision-making technique and a system-development context spanning data, models, infrastructure, and serving.
How deep it goes
Traces requirements, data exploration, prototyping, evaluation, and deployment, while stopping short of mathematical or algorithm-internal detail.
Examples
Uses a customer-support chatbot that recognizes natural-language intent and matches questions with relevant answers.
What you can do after
Build system designs, understand existing designs, and extend machine-learning infrastructure without building every component from scratch.

Read how Designing Deep Learning Systems: A guide for software engineers explains machine learning →

Deep Learning with PyTorch, Second Edition

A class of methods in which systems learn patterns from data to perform tasks without being explicitly programmed with task rules.

How it is used
Frames data-driven function fitting as the mechanism by which systems adapt to tasks and produce outputs for new inputs.
How deep it goes
Explains parameter optimization, gradients, validation, and PyTorch training mechanics while deferring more powerful neural-network applications.
Examples
Uses noisy-data fitting, a thermometer problem, and Kepler’s model-selection story to make data-driven modeling concrete.
What you can do after
Build a simple learning algorithm, optimize parameters with PyTorch, and evaluate generalization using validation data.

Read how Deep Learning with PyTorch, Second Edition explains machine learning →

Data Storytelling with Altair and AI

Machine learning is a subfield of artificial intelligence in which computers learn patterns from examples or data to improve performance and make predictions or decisions.

How it is used
Frames machine learning as the practical foundation for understanding deep learning and generative AI.
How deep it goes
Explains labeled data, feature extraction, training, parameter optimization, and prediction, without developing implementation details.
Examples
Uses fashion assessment, image classification, sentiment analysis, spam detection, and speech recognition as concrete applications.
What you can do after
Apply machine learning by preparing labeled examples, extracting features, training a model, and predicting unseen cases.

Read how Data Storytelling with Altair and AI explains machine learning →

Learn AI-Assisted Python Programming, Second Edition

Machine learning is a field of artificial intelligence in which systems infer patterns from data to make predictions or decisions.

How it is used
Uses labeled examples and extracted features to predict categories for new objects.
How deep it goes
Defines supervised learning and demonstrates feature-based prediction, without covering model internals or advanced methods.
Examples
Uses authorship identification with book texts, author labels, writing features, and book signatures.
What you can do after
Apply labeled examples and measurable features to predict the category of a new object.

Read how Learn AI-Assisted Python Programming, Second Edition explains machine learning →

WordPress Power Toolkit: Harness AI to build next-level websites

Machine learning is a branch of artificial intelligence in which computational systems learn patterns or behaviors from data rather than being explicitly programmed for every outcome.

How it is used
Explains how artificial intelligence can pursue goals by learning behaviors and strategies rather than relying entirely on direct programming.
How deep it goes
Defines self-teaching and illustrates it through Go, but stops short of explaining training procedures or algorithm internals.
Examples
Uses AlphaGo Zero’s rule-only training and championship Go matches to demonstrate original, effective machine strategies.
What you can do after
Supports understanding how systems can discover effective strategies that humans did not explicitly provide.

Read how WordPress Power Toolkit: Harness AI to build next-level websites explains machine learning →

Introduction to Generative AI, Second Edition

Machine learning is the computational use of algorithms that learn patterns from data to improve performance on tasks.

How it is used
Frames machine learning as the general method behind useful task improvement and natural language processing.
How deep it goes
Defines three learning paradigms and gives historical context, but stops short of implementation details or mathematical treatment.
Examples
Uses language translation, the Turing test, ELIZA, and the three-way learning comparison as concrete illustrations.
What you can do after
Evaluate/understand how learning signals shape tasks and why conversational fluency does not prove reasoning.

Read how Introduction to Generative AI, Second Edition explains machine learning →

Web Components in Action

Machine learning is a field in which computational models learn patterns from data to perform tasks such as classification or prediction.

How it is used
Deploys trained models in browser features such as hand tracking while hiding implementation complexity behind Web Components.
How deep it goes
Defines training and browser inference, then stops short of detailed model internals.
Examples
Uses hand-image recognition, a camera-based hand tracker, and a Web Harp integration.
What you can do after
Apply it via TensorFlow.js and Handtrack.js to add browser-based hand tracking behind a custom Web Component.

Read how Web Components in Action explains machine learning →

Cyber Threat Hunting

A machine learning algorithm is a computational procedure that learns from input data to generate a model or produce data-driven outputs.

How it is used
Deploys machine learning to help threat hunters find anomalous network activity beyond explicit detection logic.
How deep it goes
Explains algorithms, models, unsupervised learning, data preparation, features, and interpretation while stopping short of advanced data-scientist training.
Examples
Uses network connection events, beaconing to trusted destinations, Dropbox traffic, exchanged bytes, and TLS cipher suites as examples.
What you can do after
Apply it via existing Python libraries to prepare data, build unsupervised models, and interpret anomalies during threat hunting.

Read how Cyber Threat Hunting explains machine learning →

Causal Inference for Data Science

Machine learning is a field of methods that enables computers to learn patterns or predictive relationships from data.

How it is used
Frames machine learning as a prediction tool for forecasting and automated classification, distinct from causal explanation.
How deep it goes
Defines supervised prediction, contrasts it with causation, and reaches adjustment formulas and cross-fitting without deriving model internals.
Examples
Uses illness prediction, hospital-visit frequency, sales forecasting, spam detection, image recognition, language translation, and autonomous vehicles.
What you can do after
Evaluate when predictive models fit a task, recognize causal limitations, and combine models with adjustment formulas when confounders are numerous.

Read how Causal Inference for Data Science explains machine learning →

Knowledge Graphs and LLMs in Action

Machine learning is a field of computing in which models learn patterns from data to make predictions or decisions.

How it is used
Supports intelligent advisory systems by extracting predictions, relationships, and actions from knowledge graphs.
How deep it goes
Explains graph representations, task categories, embeddings, and trade-offs, but does not provide implementation code or detailed algorithms.
Examples
Uses drug repurposing, personalized treatment planning, anomalous-node detection, recommendations, and relationship prediction as applications.
What you can do after
Evaluate graph-based ML tasks and choose representations, features, and algorithms for interconnected real-world data.

Read how Knowledge Graphs and LLMs in Action explains machine learning →

Graph Algorithms for Data Science

Machine learning is a computational approach in which models learn patterns from data to perform predictions or other tasks.

How it is used
Uses graph positions and roles to increase the predictive capability of machine learning models.
How deep it goes
Connects graph structure to prediction but stops short of explaining model training or implementation.
Examples
Uses Facebook and Twitter-style user networks, follows, likes, communities, and influential nodes as examples.
What you can do after
Evaluate how graph-derived positions and roles can provide information for predictive models.

Read how Graph Algorithms for Data Science explains machine learning →

Generative AI in Action

A scientific discipline in which computers learn patterns from data to make predictions or decisions without being explicitly programmed for each task.

How it is used
Provides the foundational toolkit for understanding data engineering, model training, inference, and generative AI applications.
How deep it goes
Defines learning, predictions, and decisions, but defers detailed machine-learning mechanisms and mathematics.
Examples
Uses spam filtering and e-commerce recommendation systems to illustrate data-driven prediction and decision-making.
What you can do after
Evaluate/understand how learned data patterns support predictions, decisions, and generative AI applications.

Read how Generative AI in Action explains machine learning →

Build Your Own Robot: Using Python, CRICKIT, and Raspberry Pi

Machine learning is the computational practice of fitting models to data so they can make predictions or decisions without being explicitly programmed for every case.

How it is used
Deploys image and sensor inputs to detect environmental events and drive robotic actions.
How deep it goes
Explains trained cascade classifiers and demonstrates a pretrained model, stopping short of broader learning algorithms or mathematical treatment.
Examples
Uses OpenCV face detection, Raspberry Pi camera images, pretrained XML models, and a servo-driven face-following camera.
What you can do after
Builds a face-detection pipeline with OpenCV and connects detected movement to robotic camera control.

Read how Build Your Own Robot: Using Python, CRICKIT, and Raspberry Pi explains machine learning →

Statistics Every Programmer Needs

Machine learning is the development of computational models that learn patterns or decisions from data rather than relying only on explicitly programmed rules.

How it is used
Supports prediction, classification, pattern discovery, and data-informed decisions across practical analytical workflows.
How deep it goes
Defines learning, statistical foundations, and model families but stops short of algorithms, formulas, validation procedures, or implementation steps.
Examples
Uses regression, linear and logistic models, decision trees, random forests, forecasting, and classification as representative applications.
What you can do after
Apply machine-learning models through Python libraries to support prediction, classification, interpretation, and practical decision-making.

Read how Statistics Every Programmer Needs explains machine learning →

Grokking Statistics

The field and set of methods for learning predictive or descriptive patterns from data.

How it is used
Deploys data-driven algorithms primarily for predictive modeling and performance on new data.
How deep it goes
Contrasts prediction, tuning, and black-box performance with statistical explanation without detailing algorithm internals.
Examples
Uses labeled training and test datasets, regression, model validation, and examples involving bias and generalization.
What you can do after
Evaluate predictive models on unseen labeled data and reason about generalization, performance, explainability, and bias.

Read how Grokking Statistics explains machine learning →

Financial AI in Practice

A computational approach in which systems learn patterns from data to make predictions or decisions.

How it is used
Finds patterns in financial data for predictions and decisions, and filters users before more costly LLM analysis.
How deep it goes
Explains pattern discovery and a filtering role, but does not specify training algorithms or model internals.
Examples
Uses credit scoring, fraud detection, investment news analysis, and churn-risk screening as financial applications.
What you can do after
Operate a traditional ML churn score to filter high-risk users before LLM-based analysis.

Read how Financial AI in Practice explains machine learning →

Machines that Think

A field of artificial intelligence in which systems improve their predictions or generated outputs by learning patterns from data rather than relying only on explicitly programmed rules.

How it is used
Frames machine learning as a route beyond fixed rules for handling uncertain, complex situations.
How deep it goes
Defines learning as updating internal settings from data and probabilities, but gives no algorithmic or mathematical detail.
What you can do after
Understand how data-driven prediction differs from explicit rules and from human learning mechanisms.

Read how Machines that Think explains machine learning →

Also covered in

What to know first

Related concepts

Questions about machine learning

What is machine learning?

Machine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.

Which Manning books explain machine learning?

Optimization Algorithms: AI techniques for design, planning, and control problems, Build AI Drug Discovery Pipelines, Inside AI, Grokking Machine Learning, Natural Language Processing in Action, Second Edition, Grokking Machine Learning, Second Edition, Deep Learning with R, Third Edition, Demand Forecasting Best Practices and 41 more.

What should I know before learning machine learning?

Manning books that teach machine learning usually introduce language model, artificial intelligence, Python and deep learning first.

What kinds of machine learning are there?

Kinds of machine learning include language model, supervised learning, deep learning and reinforcement learning.

What is machine learning a kind of?

Machine learning is a kind of artificial intelligence. Artificial intelligence is the field of computing concerned with creating systems that perform tasks associated with intelligence, such as learning, reasoning, perception, and decision-making.

What is machine learning also called?

Machine learning is also called ML.

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.