Special Session On Auto ML At IJCNN 2019

Budapest, Hungary on July 14-19, 2019

Machine learning has achieved great successes in online advertising, recommender systems, financial market analysis, computer vision, computational linguistics, bioinformatics and many other fields. However, its success crucially relies on human machine learning experts involved in all systems design stages. Commonly, humans take critical decisions that include: converting real world problems into machine learning problems, data collection, feature engineering, selecting or designing model architectures, tuning model hyper-parameters, evaluating model performance, deploying on-line systems, and so on. The complexity of these tasks, which is often beyond non-experts, together with the rapid growth of machine learning applications, have motivated 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 (Automatic Machine Learning).

We are organizing a special session on Automatic Machine Learning (AutoML) collocated with the International Joint Conference on Neural Networks 2019 (IJCNN2019) that aims to compile the latest progress on this topic. The scope of the proposed special session covers all aspects of automatic machine learning, with special interest in supervised learning. Topics of interest include, but are not restricted, to:

  • Model selection, hyper-parameter optimization, and model search
  • Neural architecture search
  • Meta learning and transfer learning
  • Automatic feature extraction / construction
  • Automatic generation of workflows / workflow reuse
  • Automatic problem "ingestion" (from raw data and miscellaneous formats)
  • Automatic feature transformation to match algorithm requirements
  • Automatic detection and handling of skewed data and/or missing values
  • Automatic acquisition of new data (active learning, experimental design)
  • Automatic report writing (providing insight on automatic data analysis)
  • Automatic selection of evaluation metrics / validation procedures
  • Automatic selection of algorithms under time/space/power constraints
  • Automatic prediction post-processing and calibration
  • Automatic leakage detection
  • Automatic inference and differentiation
  • User interfaces and human-in-the-loop approaches for AutoML
  • Automatic machine learning in the life long learning setting.
  • Automatic model generation for evolving data streams.
  • AutoML in the presence of concept drift.
  • Hyper-heuristics
  • Automatic hybridization of techniques
  • Automatic operator creation
  • Automatic heuristic generation
  • Explainable machine learning


Submissions will be handled by the INNS-IJCNN2019 submission system, and will be peer reviewed as any other IJCNN2019 submission. For information on the preparation and submission of papers please refer to the following website:

Make sure to select: S12. Automatic Machine Learning as main research topic in the submission system!

The deadline for paper submission is December 15, 2019.

Organizing team

Hugo Jair Escalante (INAOE, Mexico, ChaLearn, USA), Nelishia Pillay (University of Pretoria, South Africa), Isabelle Guyon (Universté Paris-Saclay, France ChaLearn, USA), Rong Qu (University of Nottingham, UK), Wei-Wei Tu (4Paradigm Inc., China)


Please refer any question to: