KDD 2026 Tutorial · Jeju, Korea

Taming Structured Data Foundation Models with AutoML: A Hands-On Guide

A 3-hour hands-on tutorial teaching attendees how to operationalize Foundation Models for Structured Data (FM4SD) using AutoML across temporal and tabular modalities, with a highlight on Chronos-2.

View Tutorial Outline Open Notebooks Meet the Tutors Chronos-2 Paper

Notebooks

Hands-on Colab notebooks covering AutoGluon, Chronos-2, and tabular foundation models from setup to advanced topics.

Browse notebooks →

Tutorial Paper

The accepted KDD’26 tutorial paper describing the motivation, outline, and prior presentations.

Coming soon

AutoGluon

Open-source AutoML framework powering this tutorial. State-of-the-art on tabular and time-series benchmarks.

auto.gluon.ai →

Chronos-2

New multivariate zero-shot time-series foundation model from AWS. Extends Chronos to universal forecasting.

arXiv preprint →

About This Tutorial

Structured data drives enterprise decision-making, yet building predictive pipelines for time-series and tabular modalities requires intensive feature engineering, model selection, and other “tricks of the trade”. AutoGluon, an open-source AutoML system, automates this through multi-layer stack ensembling, providing a unified API that reliably achieves state-of-the-art accuracy across time series and tabular data.

Concurrently, Foundation Models for Structured Data (FM4SD) have emerged to push the boundaries of predictive performance, enabling both powerful zero-shot inference and efficient fine-tuning. Specifically, Chronos-2 delivers state-of-the-art multivariate forecasting for time series, while a rapidly growing ecosystem of tabular foundation models (e.g., TabPFN, Mitra, TabICL) transforms how practitioners execute classification and regression tasks.

This system-focused tutorial teaches attendees how to operationalize FM4SDs using AutoML across temporal and tabular modalities. We guide participants through a technical progression: establishing AutoGluon as the orchestration framework, detailing the internal mechanics of Chronos-2 and various tabular FMs, and unifying them into production-ready pipelines.

What attendees will do: Execute Colab notebooks to tokenize multivariate time series with Chronos-2 and process structured features utilizing diverse tabular foundation models. Participants will bypass standard data formatting bottlenecks, configure AutoML parameters, and fuse fine-tuned foundation models with classical ML baselines to maximize accuracy.

Key Topics

AutoGluonChronos-2TabPFNMitraTabICLTime SeriesTabular DataFM4SDAutoMLStack EnsemblingZero-shot Forecasting

Requirements

  • Laptop with a modern web browser
  • Google Colab (free tier with CPU/T4 GPU)
  • Basic Python and ML familiarity

No GPU requiredFree tier sufficient

Tutorial Outline (3 hours)

The tutorial guides attendees from foundational AutoML principles to the deployment of cutting-edge FM4SD engines.

Part 1 Modern Toolkit 30 min Part 2 AutoGluon Fundamentals 60 min Part 3 Foundation Models Deep Dive 60 min Part 4 Q&A & Wrap-up 30 min Numbered notebooks map to the KDD tutorial flow and run on free-tier Google Colab.
Part 1 — The Modern Data Scientist’s Toolkit (30 min)

Overview of structured data challenges (tabular & time series). Introduction to AutoGluon as the unified AutoML framework. Quick setup check to ensure all attendees can initialize the libraries in Colab.

Tabular dataTime seriesAutoGluon introSetup

Part 2 — AutoGluon Fundamentals (1 hour)

The philosophy of multi-layer stacked ensembling vs. hyperparameter optimization. Coding Exercise: Attendees build base TabularPredictor and TimeSeriesPredictor pipelines, experiencing the high-level API for different modalities.

Stack EnsemblingTabularPredictorTimeSeriesPredictorCoding Exercise

Part 3 — Foundation Models for Structured Data Under the Hood (1 hour)

Chronos-2 (Time Series): Understanding time series tokenization, group attention, and multivariate zero-shot forecasting. Tabular Foundation Models: Understanding TabPFN, Mitra, and TabICL. Handling complex covariates, defining custom evaluation metrics, and optimizing inference speeds for AutoML systems of tabular foundation models.

Chronos-2TokenizationGroup AttentionTabPFNMitraTabICL

Part 4 — Q&A and Wrap-up (30 min)

Best practices for deploying AutoGluon and FM4SDs effectively. Common pitfalls in data formatting and evaluation. “Bring Your Own Data” (BYOD) challenge: Attendees upload a small CSV from their own domain for 1-on-1 architecture advice from tutors.

BYOD ChallengeBest Practices1-on-1 Advice

Hands-On Colab Notebooks

Each technical section culminates in a guided API coding exercise. Open any notebook directly in Google Colab using the links below.

Notebook 1 — Setup & Introduction

Install AutoGluon, verify Colab environment, introduce the tutorial datasets and learning objectives.

Colab link coming soon

Notebook 2 — AutoGluon Tabular

Build a TabularPredictor from scratch. Experience multi-layer stack ensembling with a few lines of code.

Colab link coming soon

Notebook 3 — AutoGluon Time Series

Build a TimeSeriesPredictor pipeline. Explore probabilistic forecasting and model leaderboards.

Colab link coming soon

Notebook 4 — Chronos-2 Deep Dive

Tokenize multivariate time series with Chronos-2. Run zero-shot multivariate forecasting and interpret group-attention patterns.

Colab link coming soon

Notebook 5 — Tabular Foundation Models

Compare TabPFN, Mitra, and TabICL on classification and regression tasks. Fine-tune and integrate with AutoGluon ensembles.

Colab link coming soon

Notebook 6 — End-to-End Pipeline

Fuse fine-tuned foundation models with classical ML baselines. Deploy a production-ready AutoML pipeline for structured data.

Colab link coming soon
Note: All notebooks run on Google Colab free tier (CPU or T4 GPU). Datasets are hosted on Hugging Face / AWS S3. No local setup required.

What Participants Learn

This tutorial provides a complete technical journey from foundational AutoML principles to cutting-edge foundation model deployment for structured data.

SETUPENSEMBLINGFORECASTINGFINE-TUNINGDEPLOYMENTBYOD

  • Understand the structured data challenge landscape
  • Use AutoGluon’s unified API for tabular and time-series tasks
  • Apply Chronos-2 for zero-shot multivariate forecasting
  • Leverage TabPFN, Mitra, and TabICL for tabular tasks
  • Fine-tune FM4SDs and fuse with classical baselines
  • Evaluate with custom metrics and uncertainty quantification
BYOD Challenge: Because AutoGluon requires only three lines of code (init, fit, predict), we dedicate the final 15 minutes to a “Bring Your Own Data” challenge where tutors provide 1-on-1 architecture advice on real participant data.
Time Series multivariate Tabular classification/regression AutoGluon Multi-layer stack ensembling Chronos-2 time series TabPFN tabular Mitra tabular Production Pipeline Classical ML Baselines LightGBM, CatBoost, XGBoost… Fused with FM4SDs for max accuracy

Tutorial Organizers

Prior Labs
In-Person Presenter

Nick Erickson

Prior Labs

Lead developer and maintainer of AutoGluon and TabArena. Research scientist focused on AutoML and foundation models for structured data.

nick@priorlabs.ai

Amazon Web Services
In-Person Presenter

Yuyang (Bernie) Wang

Amazon Web Services

Principal Scientist at AWS, where he aims to democratize advanced AI/ML capabilities, making them accessible to practitioners across diverse domains.

yuyawang@amazon.com

Amazon Web Services

Boran Han

Amazon Web Services

Senior applied scientist working on AutoML, foundation models for structured data, and agentic frameworks. Core contributor to AutoGluon and MLZero.

boranhan@amazon.com

Amazon Web Services

Abdul Fatir Ansari

Amazon Web Services

Lead author of the Chronos series. Researcher working on deep learning for time series, probabilistic forecasting, and foundation models.

ansarnd@amazon.com

Amazon Web Services

Oleksandr Shchur

Amazon Web Services

Lead author of AutoGluon-TimeSeries. Researcher specializing in probabilistic time series forecasting and AutoML for temporal data.

shchuro@amazon.com

Amazon Web Services

Xiyuan Zhang

Amazon Web Services

Lead author of Mitra and co-author of the Chronos series. Applied scientist working on foundation models for tabular and time series data.

xiyuanz@amazon.com

Amazon Web Services

Shuai Zhang

Amazon Web Services

Senior Applied Scientist at AWS. His research focuses on LLMs, retrieval-augmented generation, reinforcement learning, and machine learning systems.

shuaizs@amazon.com

Amazon Web Services

Haoyang Fang

Amazon Web Services

Applied scientist working on agentic frameworks to enhance RL performance. Develops AutoGluon Multimodal and leads MLZero, a multi-agent system for AutoML.

haoyfang@amazon.com

Amazon Web Services

Danielle Maddix

Amazon Web Services

Senior Applied Scientist at AWS AI. Danielle has worked on statistical and deep learning foundation models for structured data. She also leads the sciML research initiative.

dmmaddix@amazon.com

Amazon Web Services

Michael Bohlke-Schneider

Amazon Web Services

Applied Science Manager at AWS. At AWS, Michael works on machine learning and forecasting, with a focus on foundation models for structured data and AutoML.

bohlkem@amazon.com

Prior Presentations

KDD 2020

AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data

Virtual — August 2020

Introduced AutoGluon’s core ensembling architecture for tabular data. Focused on robust, high-accuracy AutoML without manual tuning.

KDD 2022

Multimodal AutoML for Image, Text and Tabular Data

Washington, DC — August 2022 — ~200 attendees

Extended AutoGluon to multi-modal settings including text and image data alongside tabular modalities.

What’s new in 2026: While previous tutorials introduced AutoGluon’s core ensembling and multi-modal capabilities, the 2026 tutorial represents a major paradigm shift by adding modern Foundation Models for Structured Data (FM4SD) and introducing time series forecasting with Chronos-2—a distinct temporal modality not covered in previous tutorials.

Key Papers & Resources

Paper Venue Link
AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data — Erickson et al. AutoML Workshop @ ICML 2020 arXiv
AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecasting — Shchur et al. AutoML Conference 2023 arXiv
Chronos: Learning the Language of Time Series — Ansari et al. TMLR 2024 arXiv
Chronos-2: From Univariate to Universal Forecasting — Ansari et al. arXiv 2025 arXiv
Accurate predictions on small data with a tabular foundation model (TabPFN) — Hollmann et al. Nature 2025 Nature
Mitra: Mixed Synthetic Priors for Enhancing Tabular Foundation Models — Zhang et al. NeurIPS 2025 arXiv
TabICL: A Tabular Foundation Model for In-Context Learning on Large Data — Qu et al. ICML 2025 arXiv

Societal Impact

Improved tabular and time series modeling impacts critical resource allocation in healthcare, finance, and logistics. While mastering tools like AutoGluon and its engines democratizes powerful predictive capabilities, it carries the risk of automation bias—where users blindly trust ML outputs.

This tutorial addresses this by emphasizing rigorous data science practices such as probabilistic forecasting and uncertainty quantification, ensuring attendees learn to evaluate model uncertainty and apply appropriate skepticism to automated predictions.