Auto-WEKA
{{Short description|Automated machine learning system}}
Auto-WEKA is an automated machine learning system based on Weka by Chris Thornton, Frank Hutter, Holger H. Hoos and Kevin Leyton-Brown.{{Cite web|url=https://doi.org/10.1145/2487575.2487629|title=Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms|first1=Chris|last1=Thornton|first2=Frank|last2=Hutter|first3=Holger H.|last3=Hoos|first4=Kevin|last4=Leyton-Brown|date=August 11, 2013|publisher=Association for Computing Machinery|pages=847–855|via=ACM Digital Library|doi=10.1145/2487575.2487629}} An extended version was published as Auto-WEKA 2.0.{{Cite journal|url=http://jmlr.org/papers/v18/16-261.html|title=Auto-WEKA 2.0: Automatic model selection and hyperparameter optimization in WEKA|first1=Lars|last1=Kotthoff|first2=Chris|last2=Thornton|first3=Holger H.|last3=Hoos|first4=Frank|last4=Hutter|first5=Kevin|last5=Leyton-Brown|date=August 12, 2017|journal=Journal of Machine Learning Research|volume=18|issue=25|pages=1–5|via=jmlr.org}} Auto-WEKA was named the first prominent AutoML system in a neutral comparison study.{{cite journal |last1=Gijsbers |first1=Pieter |last2=Bueno |first2=Marcos L. P. |title=AMLB: an AutoML Benchmark |journal=Journal of Machine Learning Research |date=2024 |volume=25 |page=6 |arxiv=2207.12560 |url=http://jmlr.org/papers/v25/22-0493.html}}
It received the test-of-time award of the SIGKDD conference in 2023.{{Cite web|url=https://www.kdd.org/kdd2023/awards/index.html|title=KDD 2023 - Awards|website=kdd.org}}
Description
Auto-WEKA introduced the Combined Algorithm Selection and Hyperparameter optimization (CASH) problem, that extends both the Algorithm selection problem and the Hyperparameter optimization problem, by searching for the best algorithm and also its hyperparameters for a given dataset. Baratchi et al. state that "[T]he real power of AutoML was unlocked through the definition of the combined algorithm selection and hyperparameter optimisation problem".{{cite journal |last1=Baratchi |first1=Mitra |last2=Wang |first2=Can |last3=Limmer |first3=Steffen |last4=van Rijn |first4=Jan N. |last5=Hoos |first5=Holger |last6=Bäck |first6=Thomas |last7=Olhofer |first7=Thomas |title=Automated machine learning: past, present and future |journal=Artificial Intelligence Review |volume=57 |issue=5 |page=2 |doi=10.1007/s10462-024-10726-1 |date=2024|doi-access=free }}
The CASH for formalism was picked up and also extended by later AutoML systems and methods such as Auto-sklearn,{{cite conference |last1=Feurer | first1=Matthias | last2=Klein | first2=Aaron | last3=Eggensperger | first3=Katharina | last4=Springenberg | first4=Jost Tobias | last5=Blum | first5=Manuel | last6=Hutter | first6=Frank | title=Efficient and Robust Automated Machine Learning | url=http://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine-learning | book-title=Advances in Neural Information Processing Systems | volume=28 |date=2015 }}
ATM,{{cite conference |last1=Swearingen |first1=Thomas |last2=Drevo |first2=Will |last3=Cyphers |first3=Benett |last4=Cuesta-Infante |first4=Alfredo |last5=Ross |first5=Arun |last6=Veeramachaneni |first6=Kalyan |title=ATM: A distributed, collaborative, scalable system for automated machine learning |book-title=2017 IEEE International Conference on Big Data (Big Data)|doi=10.1109/BigData.2017.8257923 | url=https://ieeexplore.ieee.org/document/8257923 |date=2017|url-access=subscription }} AutoPrognosis,{{cite conference | last1=Alaa | first1=Ahmed M. | last2=van der Schaar | first2=Mihaela | title=AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning | book-title=Proceedings of the 35th International Conference on Machine Learning | date=2018 | url=https://proceedings.mlr.press/v80/alaa18b.html }} MCPS,{{cite journal |last1=Salvador |first1=Manuel Martin |last2=Budka |first2=Marcin |last3=Gabrys |first3=Bogdan |title=Automatic Composition and Optimization of Multicomponent Predictive Systems With an Extended Auto-WEKA |journal=IEEE Transactions on Automation Science and Engineering |date=2019 |volume=16 |issue=2 |pages=946–959 |doi=10.1109/TASE.2018.2876430 |url=https://ieeexplore.ieee.org/document/8550732|arxiv=1612.08789 }} MOSAIC,{{cite conference | last1=Rakotoarison | first1=Herilalaina | last2=Schoenauer | first2=Marc | last3=Sebag | first3=Michèle | title=Automated Machine Learning with Monte-Carlo Tree Search
| book-title=Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence | date=2019 | url=https://doi.org/10.24963/ijcai.2019/457 | doi=10.24963/ijcai.2019/457 | arxiv=1906.00170 }} naive AutoML{{cite journal |last1=Mohr |first1=Felix |last2=Wever |first2=Marcel | title=Naive automated machine learning |journal=Machine Learning |date=2023 |volume=112 |issue=4 |pages=1131–1170 |doi=10.1007/s10994-022-06200-0 |doi-access=free |arxiv=2111.14514 }} and ADMM.{{cite conference | last1=Liu | first1=Sijia | last2=Ram | first2=Parikshit | last3=Vijaykeerthy | first3=Deepak | last4=Bouneffouf | first4=Djallel | last5=Bramble | first5=Gregory | last6=Samulowitz | first6=Horst | last7=Wang | first7=Dakuo | last8=Conn | first8=Andrew | last9=Gray | first9=Alexander | title=An ADMM based framework for automl pipeline configuration | book-title=Proceedings of the AAAI Conference on Artificial Intelligence | volume=34 | number=4 | url=https://doi.org/10.1609/aaai.v34i04.5926 | doi=10.1609/aaai.v34i04.5926 |date=2020 | arxiv=1905.00424 }}
References
{{reflist}}
Category:Data mining and machine learning software
{{software-stub}}
{{machine-learning-stub}}