Auto-WEKA 2.0: Automatic model selection and hyperparameter optimization in WEKA

By Lars Kotthoff, Chris J. Thornton, Holger H. Hoos, Frank Hutter, Kevin Leyton-Brown. 2017

652 citations recorded by the source corpus.

Subfields: Neural Architecture Search & AutoML

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Neural Architecture Search & AutoML

Citations in the corpus

Automated Machine Learning: State-of-The-Art and Open Challenges (2019, background): “For example, Auto-Weka(6) is considered as the first and pioneer machine learning automation framework [66].”
Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “The sister package Auto-WEKA (Thornton et al., 2013; Kotthoff et al., 2016) provides very similar functionality for the WEKA library.”
Auto-Keras: An Efficient Neural Architecture Search System (2018, methodology): “For example, it has been used in hyperparameter tuning for machine learning models [13, 15, 17, 35], in which Bayesian optimization searches among different combinations of hyperparameters.”
A System for Massively Parallel Hyperparameter Tuning (2018, methodology): “AutoWEKA (Kotthoff et al., 2017) and AutoSklearn (Feurer et al., 2015) are two established single-machine, single-user systems for hyperparameter optimization.”
Efficient Neural Architecture Search with Network Morphism (2018, methodology): “For example, it has been used in hyperparameter tuning for machine learning models [26, 14, 9, 11], each observation of which involves the training and testing of a machine learning model, which is very similar to the NAS problem.”
Human-AI Collaboration in Data Science (2019, neutral): “databases. There are also techniques that try to engineer new features for a target ML model using reinforcement learning [37] and guided search based on past experiences [49]. Tools such as AutoWEKA [41, 66], Auto-sklearn [15], and TPOT [13, 52] all automate the model selection, hyperparameter opti”
Automated Algorithm Selection: Survey and Perspectives (2018, background): “Realistic algorithm configuration scenarios typically involve tens to hundreds of parameters (see, e.g., Hutter et al., 2009, 2011; López-Ibáñez et al., 2016; Ansótegui et al., 2015; Thornton et al., 2013; Kotthoff et al., 2017).”
AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning (2018, methodology): “W EKA (Kotthoff et al., 2016), A UTO - SKLEARN (Feurer et al., 2015), and TPOT (Olson & Moore, 2016).”
Automatic Model Monitoring for Data Streams (2019, methodology): “Early research on autoML has mainly focused on model selection and hyperparameter tuning [7, 17, 25].”
Ten quick tips for machine learning in computational biology (2017, methodology): “These packages include Auto-Sklearn [35], Auto-Weka [36], TPOT [37], and PennAI [38].”
Auto-Keras: Efficient Neural Architecture Search with Network Morphism (2018, methodology): “For example, it has been used in hyperparameter tuning for machine learning models (Thornton et al., 2013; Kotthoff et al., 2016; Feurer et al., 2015; Hutter et al., 2011), in which Bayesian optimization searches among different combinations of hyperparameters.”
ML-Plan: Automated machine learning via hierarchical planning (2018, methodology): “Auto-WEKA (Thornton et al. 2013; Kotthoff et al. 2017) and auto-sklearn Feurer et al. (2015) are the main representatives for solving AutoML by so-called sequential parameter optimization.”
Democratizing Data Science through Interactive Curation of ML Pipelines (2019, background): “Most noticeable, TuPAQ [35, 37], Hyper-band [22] and the various Bayesian Optimization approaches [11, 17, 40] all have the goal to automatically determine the best model family (e.g., SVM vs Linear regression) or parameters for a given algorithm (e.g., step-size, kernel, etc.).”
Bandit-Based Automated Machine Learning (2018, background): “[10] describe the new version of Auto-WEKA, the AutoWeka 2.”
AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data (2020, methodology): “Most existing AutoML frameworks instead just apply standard feedforward architectures to one-hot encoded data (Kotthoff et al., 2017; Pandey, 2019).”