Call for Papers

The AutoML Workshop @ ICML 2015
Lille, France, July 11, 2015

Important Dates:
- Submission deadline: 1 May, 2015, 11:59pm UTC-12
- Notification of acceptance:10 May, 2015
- Submission deadline (late breaking papers): 8 June, 2015, 11:59pm UTC-12
- Notification of acceptance (late breaking papers): 18 June, 2015

Workshop Overview:
Machine learning has achieved considerable successes in recent years, but these successes crucially rely on human machine learning experts, who select appropriate features, workflows, machine learning paradigms, algorithms, and their hyperparameters. As the complexity of these tasks is often beyond non-experts, the rapid growth of machine learning applications has created a demand for off-the-shelf machine learning methods that can be used easily and without expert knowledge. We call the resulting research area that targets progressive automation of machine learning AutoML.

AutoML aims to automate many different stages of the machine learning process such as:
  • Model selection, hyper-parameter optimization, and model search
  • Representation learning and automatic feature extraction / construction
  • Reusable workflows and automatic generation of workflows
  • Meta learning and transfer learning
  • Automatic problem "ingestion" (from raw data and miscellaneous formats)
  • Feature coding/transformation to match requirements of different learning algorithms
  • Automatically detecting and handling skewed data and/or missing values
  • Automatic leakage detection
  • Matching problems to methods/algorithms (beyond regression and classification)
  • Automatic acquisition of new data (active learning, experimental design)
  • Automatic report writing (providing insight on the data analysis performed automatically)
  • User interfaces for AutoML (e.g., “Turbo Tax for Machine Learning”)
  • Automatic inference and differentiation
  • Automatic selection of evaluation metrics
  • Automatic creation of appropriately sized and stratified train, validation, and test sets
  • Parameterless, robust algorithms
  • Automatic selection of algorithms to satisfy time/space/power constraints at train-time or at run-time
  • Run-time protection wrappers to detect data shift and other causes of prediction failure
We encourage contributions in any of these areas; for details on submission please see the submission site.

Invited speakers:
  • Rich Caruana: Open Research Problems in AutoML
  • David Duvenaud: Automatically Constructing Models, and Automatically Explaining them, too
  • Matt Hoffman: Bandits and Bayesian optimization for AutoML
  • Jürgen Schmidhuber: Recursive Self-Improvement
  • Michele Sebag: Algorithm Recommendation as Collaborative Filtering
  • Joaquin Vanschoren: OpenML: Networked Machine Learning
Organizers: Frank Hutter, Balázs Kégl, Isabelle Guyon, Hugo Larochelle, and Evelyne Viegas