AutoML

Mitarbeiter: Marius Lindauer, Theresa Eimer, Difan Deng
Introduction

To use machine learning (ML), users have to choose between many design options: (i) ML algorithms (ii) pre-processing techniques, (iii) post-processing techniques, (iv) hyperparameter settings, (v) architectures of neural networks and so on. These design decisions are often responsible whether ML systems return random predictions or achieve state-of-the-art performance. Unfortunately, even for ML-experts it is a tedious and error-prone task and thus it is not easy to make these decisions efficiently.

Automated machine learning (AutoML) addresses this challenge by automating the design process such that AutoML tools support users to efficiently develop new ML applications.

Recent Research Topics

Hyperparameter Optimization and Bayesian Optimization

To achieve peak-performance with an algorithm, choosing an appropriate hyperparameter configuration is crucial. Since hyperparameters are often not very intuitive for human developers, it is a tedious and error-prone task to choose these settings. Bayesian Optimization is a sample-efficient approach to find such hyperparameter configurations in an automatic way, saving human developers tremendous amounts of development time.

Neural Architecture Search

Applying deep learning to new datasets also requires to find a well-performing architecture of a deep neural network. Such an architecture influences the performance, but also other metrics, such as inference time, memory consumption etc pp. Unfortunately, it is again not obvious for human developers how to design such deep neural networks making the process fairly inefficient. Neural architecture search is an paradigma to automatically determine the best architectures for new datasets, making new applications of deep learning feasible also at larger scale.

Dynamic Algorithm Configuration

Instead of choosing the hyperparameters of an ML algorithm once, many hyperparameters have to be adapted over time. A well-known example is the learning rate of a deep neural network, which is decreased, sometimes also increased, over time. So far, these dynamic hyperparameters are controlled by a human-designed heuristic, which is often not optimal for a new dataset. Therefore, we develop new approaches for dynamic algorithm configuration, which learns from data how to adjust these on-the-fly.

Interpretability of AutoML 

A major drawback of AutoML tools is the risk that ML will be even a more mysterious black box than it ever was. Therefore, we also develop analysis tools that provide feedback to AutoML users about important insights, such as, (i) how to use AutoML tools more efficiently or (ii) which hyperparameter decisions were important to achieve the final performance. This helps ML developers to get a better understanding of why and how ML and AutoML works.

  • Conference Contributions
    • Andre Biedenkapp, H. Furkan Bozkurt, Theresa Eimer, Frank Hutter, Marius Lindauer
      Algorithm Control: Foundation of a New Meta-Algorithmic Framework
      Proceedings of the European Conference on Artificial Intelligence (ECAI), 2020
    • David Speck, André Biedenkapp, Frank Hutter, Robert Mattmüller, Marius Lindauer
      Learning Heuristic Selection with Dynamic Algorithm Configuration
      Proceedings of international workshop on Bridging the Gap Between AI Planning and Reinforcement Learning at ICAPS, June 2020
    • Theresa Eimer, Andre Biedenkapp, Frank Hutter, Marius Lindauer
      Towards Self-Paced Context Evaluations for Contextual Reinforcement Learning
      Workshop on Inductive Biases, Invariances and Generalization in Reinforcement Learning (BIG@ICML'20), July 2020
    • Gresa Shala, Andre Biedenkapp, Noor Awad, Steven Adriaensen, Marius Lindauer, Frank Hutter
      Learning Step-Size Adaptation in CMA-ES
      Proceedings of the Sixteenth International Conference on Parallel Problem Solving from Nature ({PPSN}'20), September 2020
    • Berend Denkena, Marc Dittrich, Marius Lindauer, Mainka , Lukas Stürenburg
      Using AutoML to Optimize Shape Error Prediction in Milling Processes
      Proceedings of 20th Machining Innovations Conference for Aerospace Industry (MIC), December 2020
    • Andre Biedenkapp, Raghu Rajan, Frank Hutter, Marius Lindauer
      Towards TempoRL: Learning When to Act
      Workshop on Inductive Biases, Invariances and Generalization in Reinforcement Learning (BIG@ICML'20), July 2020
    • M. Lindauer and M. Feurer and K. Eggensperger and A. Biedenkapp and F. Hutter
      Towards Assessing the Impact of Bayesian Optimization's Own Hyperparameters
      {IJCAI} 2019 {DSO} Workshop, August 2019
    • M. Feurer and K. Eggensperger and S. Falkner and M. Lindauer and F. Hutter
      Practical Automated Machine Learning for the AutoML Challenge 2018
      ICML 2018 AutoML Workshop, July 2018
    • K. Eggensperger and M. Lindauer and F. Hutter
      Neural Networks for Predicting Algorithm Runtime Distributions
      Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI’18), pp. 1442-1448, July 2018
    • A. Biedenkapp and J. Marben and M. Lindauer and F. Hutter
      CAVE: Configuration Assessment, Visualization and Evaluation
      Proceedings of the International Conference on Learning and Intelligent Optimization (LION'18), June 2018
  • Journals
    • Marius Lindauer and Frank Hutter
      Best Practices for Scientific Research on Neural Architecture Search
      Journal of Machine Learning Research, Vol. 21, pp. 1-18, December 2020
  • Book Chapters
    • Hector Mendoza and Aaron Klein and Matthias Feurer and Jost Tobias Springenberg and Matthias Urban and Michael Burkart and Max Dippel and Marius Lindauer and Frank Hutter
      Towards Automatically-Tuned Deep Neural Networks
      AutoML: Methods, Sytems, Challenges, Springer, pp. 141--156, December 2018, edited by Hutter, Frank and Kotthoff, Lars and Vanschoren, Joaquin
  • Technical Report
    • Artur Souza, Luigi Nardi, Leonardo Oliveira, Kunle Olukotun, Marius Lindauer, Frank Hutter
      Prior-guided Bayesian Optimization
      arxiv:2006.14608[cs.LG], June 2020
    • Lucas Zimmer, Marius Lindauer, Frank Hutter
      Auto-PyTorch Tabular: Multi-Fidelity MetaLearning for Efficient and Robust AutoDL
      arxiv:2006.13799[cs.LG], June 2020
    • Matthias Feurer, Katharina Eggensperger, Stefan Falkner, Marius Lindauer, Frank Hutter
      Auto-Sklearn 2.0: The Next Generation
      arXiv:2007.04074 [cs.LG], July 2020
    • Katharina Eggensperger, Kai Haase, Philipp Müller, Marius Lindauer, Frank Hutter
      Neural Model-based Optimization with Right-Censored Observations
      CoRR, ArXiv, September 2020
    • Zhengying Liu, Adrien Pavao, Zhen Xu, Sergio Escalera, Fabio Ferreira, Isabelle Guyon, Sirui Hong, Frank Hutter, Rongrong Ji, Julio Jacques, Ge Li, Marius Lindauer, Zhipeng Luo, Meysam Madadi, Thomas Nierhoff, Kangning Niu, Chunguang Pan, Danny Stoll, Sebastien Treguer, Jin Wang, Peng Wang, Chenglin Wu, Youcheng Xiong, Arbër Zela, Yang Zhang
      Winning solutions and post-challenge analyses of the ChaLearn AutoDL challenge 2019
      HAL, September 2020
    • M. Lindauer and K. Eggensperger and M. Feurer and A. Biedenkapp and J. Marben and P. M\"uller and F. Hutter
      BOAH: A Tool Suite for Multi-Fidelity Bayesian Optimization & Analysis of Hyperparameters
      arXiv:1908.06756 [cs.LG], August 2019