tabular data
In plain wordsData arranged as a table, like a spreadsheet: each row is a case — a customer, a loan, a patient — and each column a measured property. Much business and scientific data looks like this.
Data organized as a table in which rows are observations and columns are featuresfeature, n.: One piece of information about each example that a model uses, like a house's size or a pixel's color, or a pattern a model has learned to detect inside itself. of mixed types — numeric, categorical, ordinal, dates — typically drawn from databases, spreadsheets, logs or surveys. Unlike images, audio or text, tabular columns have no natural order or spatial structure, carry different units and meanings, and often include missing values, skewed distributions, high-cardinality categories and abrupt, threshold-like relations to the target. These properties suit the inductive biasinductive bias, n.: The assumptions built into a learning method that decide how it fills the gaps between the examples it was shown, such as that similar inputs get similar answers. Every learner needs some to cope with new cases. of tree ensembles: through the 2010s and early 2020s, gradient-boostedgradient boosting, n.: A method that builds a strong predictor by adding small decision trees one at a time, each trained to correct the errors that remain after all the trees before it. trees (XGBoostXGBoost, prop. n.: A fast, free software tool that makes predictions by adding up hundreds of small decision trees, each one fixing the mistakes of those before it. For years it was behind many winning entries in competitions on spreadsheet-style data., LightGBMLightGBM, prop. n.: A widely used tool for making predictions from very large tables of data, quickly and in little memory, by combining many small decision trees. Microsoft released it as free software in 2016., CatBoostCatBoost, prop. n.: A free gradient-boosting library from the tech company Yandex, known for handling categories such as city names or product types directly, without first converting them into numbers by hand.) and random forestsrandom forest, n.: A prediction method that builds hundreds of decision trees, each from a random sample of the data and features, and lets them vote. The crowd is far more reliable than any single tree. generally matched or beat deep networks on medium-sized tables, as systematic benchmarks confirmed in 2022. Dedicated architectures (TabNet, FT-Transformer and others) narrowed the gap, and since 2025 tabular foundation modelstabular foundation model, n.: A neural network pretrained on millions of made-up tables so that, given a new spreadsheet of labeled examples, it can predict the missing answers at once, without being retrained for that table. pretrained on synthetic tables have rivaled tuned boosting on small and medium datasets without per-task training. Most enterprise machine learning — credit scoring, churn, fraud, demand forecasting, clinical risk — runs on tabular data.
For tabular data, start with gradient-boosted trees; try a neural network only if you have a reason.
Origin
Tables are as old as statistics; 'tabular data' became a named machine-learning category in the late 2010s, as researchers asked why deep learning lagged on it. Ravid Shwartz-Ziv and Amitai Armon's 'Tabular Data: Deep Learning Is Not All You Need' (2022) and Léo Grinsztajn, Edouard Oyallon and Gaël Varoquaux's 'Why Do Tree-Based Models Still Outperform Deep Learning on Tabular Data?' (2022) became reference benchmarks.
Usage note
As of October 2026, whether foundation models will displace boosted trees on large tables remains open. Tables inside documents ('table understanding') and spreadsheets edited by language-model agents are related but separate problems.
Related
- XGBoostXGBoost, prop. n.: A fast, free software tool that makes predictions by adding up hundreds of small decision trees, each one fixing the mistakes of those before it. For years it was behind many winning entries in competitions on spreadsheet-style data.
- gradient boostinggradient boosting, n.: A method that builds a strong predictor by adding small decision trees one at a time, each trained to correct the errors that remain after all the trees before it.
- random forestrandom forest, n.: A prediction method that builds hundreds of decision trees, each from a random sample of the data and features, and lets them vote. The crowd is far more reliable than any single tree.
- tabular foundation modeltabular foundation model, n.: A neural network pretrained on millions of made-up tables so that, given a new spreadsheet of labeled examples, it can predict the missing answers at once, without being retrained for that table.
- feature engineeringfeature engineering, n.: Turning raw data into the inputs a model learns from, using knowledge of the problem; for example, converting a timestamp into 'hour of day' and 'is it a weekend'.
- KaggleKaggle, prop. n.: A popular website, owned by Google, where people compete to build the most accurate prediction models on shared datasets, publish their code and data, and learn data science.
- one-hot encodingone-hot encoding, n.: Turning a category such as 'red', 'green' or 'blue' into a row of zeros with a single one marking which category it is, so that a model can use it as numbers.
Sources
- Grinsztajn, L., Oyallon, E. & Varoquaux, G. (2022). Why Do Tree-Based Models Still Outperform Deep Learning on Tabular Data? NeurIPS 2022 Datasets and Benchmarks Track. arXiv:2207.08815. arxiv.org/abs/2207.08815
- Shwartz-Ziv, R. & Armon, A. (2022). Tabular Data: Deep Learning Is Not All You Need. Information Fusion, 81, 84–90.
Mentioned in
datadata, n.: Recorded facts, measurements, text, images or sounds that a computer can store and process. AI systems learn their skills from large amounts of data rather than from rules written by hand., AutoMLAutoML, n.: Software that automates the trial-and-error parts of building a machine-learning model, such as choosing the algorithm, its settings and the data-preparation steps, so less expert effort is needed., boostingboosting, n.: Building a strong predictor from many weak ones trained one after another, where each new model concentrates on the mistakes the previous ones made., CatBoostCatBoost, prop. n.: A free gradient-boosting library from the tech company Yandex, known for handling categories such as city names or product types directly, without first converting them into numbers by hand., feature engineeringfeature engineering, n.: Turning raw data into the inputs a model learns from, using knowledge of the problem; for example, converting a timestamp into 'hour of day' and 'is it a weekend'., gradient boostinggradient boosting, n.: A method that builds a strong predictor by adding small decision trees one at a time, each trained to correct the errors that remain after all the trees before it., KaggleKaggle, prop. n.: A popular website, owned by Google, where people compete to build the most accurate prediction models on shared datasets, publish their code and data, and learn data science., Leo BreimanLeo Breiman, prop. n.: An American statistician who invented random forests and bagging, decision-tree methods still in wide use, and urged statisticians to judge models by how well they predict., LightGBMLightGBM, prop. n.: A widely used tool for making predictions from very large tables of data, quickly and in little memory, by combining many small decision trees. Microsoft released it as free software in 2016., multilayer perceptronmultilayer perceptron, n.: A neural network made of several layers of artificial neurons, each connected to every neuron in the next layer. It is the simplest kind of deep network and a component of almost every modern model., one-hot encodingone-hot encoding, n.: Turning a category such as 'red', 'green' or 'blue' into a row of zeros with a single one marking which category it is, so that a model can use it as numbers., predictive AIpredictive AI, n.: AI used to forecast outcomes about people or events, such as who will repay a loan, which patient may get sicker or which machine will fail, in order to guide decisions., random forestrandom forest, n.: A prediction method that builds hundreds of decision trees, each from a random sample of the data and features, and lets them vote. The crowd is far more reliable than any single tree., XGBoostXGBoost, prop. n.: A fast, free software tool that makes predictions by adding up hundreds of small decision trees, each one fixing the mistakes of those before it. For years it was behind many winning entries in competitions on spreadsheet-style data., explainable boosting machineexplainable boosting machine, n.: A prediction method for spreadsheet-style data that draws each input's influence as its own readable graph and adds those effects up to make every prediction, usually losing little accuracy compared with opaque methods., tabular foundation modeltabular foundation model, n.: A neural network pretrained on millions of made-up tables so that, given a new spreadsheet of labeled examples, it can predict the missing answers at once, without being retrained for that table.