Towards a method for automatically selecting and configuring multi-label classification algorithms

Sá, Alex Guimarães Cardoso de; Pappa, Gisele L.; Freitas, Alex A.

doi:10.1145/3067695.3082053

Cited by 19 publications

(24 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By contrast, we propose Auto-MEKA GGP , a grammar-based genetic programming method to solve the Auto-ML task for multi-label data. Auto-MEKA GGP overcomes the major drawbacks of our previously proposed GA-Auto-MLC method [2], being able to properly handle the complex hierarchical nature of the MLC search space. It is important to point out that in this paper we focus only on algorithms and hyper-parameters (not pipelines), as the MLC search space is much bigger than the SLC search space.…”

Section: Related Workmentioning

confidence: 99%

“…This is because of the higher difficulty to learn from multi-label data, the strain to evaluate the produced MLC models [11,16], and the computational cost involved. Despite these problems, we have recently proposed the first Auto-ML method to tackle MLC [2], here referred to as GA-Auto-MLC. The method is a simple real-coded genetic algorithm (GA) that performs a search in a very large (hierarchical) search space of many different types of MLC algorithms from the MEKA framework [13].…”

Section: Introductionmentioning

confidence: 99%

“…The method is a simple real-coded genetic algorithm (GA) that performs a search in a very large (hierarchical) search space of many different types of MLC algorithms from the MEKA framework [13]. Although GA-Auto-MLC was effective in the experiments reported in [2] (with only three datasets), its solution encoding approach has two major drawbacks: it is cumbersome and it allows individuals representing impractical MLC algorithm configurations (in the sense that the MLC algorithms could have invalid configurations or take too long to run).…”

Section: Introductionmentioning

confidence: 99%

“…In order to evaluate its effectiveness, Auto-MEKA GGP was tested in 10 datasets and compared to two well-known MLC algorithms: Binary Relevance (BR) [16] and Classifier Chain (CC) [12]. Auto-MEKA GGP was also compared to GA-Auto-MLC, and all comparisons were based on a combination of several multi-label predictive accuracy measures [2,11,16]. We run two versions of Auto-MEKA GGP : a full version with our proposed grammar, and a simplified grammar version including only the components of GA-Auto-MLC.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automated Selection and Configuration of Multi-Label Classification Algorithms with Grammar-Based Genetic Programming

Sá

Freitas

Pappa

2018

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

This paper proposes Auto-MEKAGGP, an Automated Machine Learning (Auto-ML) method for Multi-Label Classification (MLC) based on the MEKA tool, which offers a number of MLC algorithms. In MLC, each example can be associated with one or more class labels, making MLC problems harder than conventional (single-label) classification problems. Hence, it is essential to select an MLC algorithm and its configuration tailored (optimized) for the input dataset. Auto-MEKAGGP addresses this problem with two key ideas. First, a large number of choices of MLC algorithms and configurations from MEKA are represented into a grammar. Second, our proposed Grammar-based Genetic Programming (GGP) method uses that grammar to search for the best MLC algorithm and configuration for the input dataset. Auto-MEKAGGP was tested in 10 datasets and compared to two well-known MLC methods, namely Binary Relevance and Classifier Chain, and also compared to GA-Auto-MLC, a genetic algorithm we recently proposed for the same task. Two versions of Auto-MEKAGGP were tested: a full version with the proposed grammar, and a simplified version where the grammar includes only the algorithmic components used by GA-Auto-MLC. Overall, the full version of Auto-MEKAGGP achieved the best predictive accuracy among all five evaluated methods, being the winner in six out of the 10 datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated Selection and Configuration of Multi-Label Classification Algorithms with Grammar-Based Genetic Programming

Sá

Freitas

Pappa

2018

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Automatic machine learning [8], or AutoML, approaches have seen a recent resurgence of interest as researchers look for ways to automatically select optimal algorithms, features, model architectures, and hyperparameters for machine learning tasks. The AutoML research community has, however, paid very little attention to multi-label classification problems, although there have been some recent efforts [25,26,33].…”

Section: Introductionmentioning

confidence: 99%

CascadeML: An Automatic Neural Network Architecture Evolution and Training Algorithm for Multi-label Classification (Best Technical Paper)

Pakrashi

Namee

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Multi-label classification is an approach which allows a datapoint to be labelled with more than one class at the same time. A common but trivial approach is to train individual binary classifiers per label, but the performance can be improved by considering associations within the labels. Like with any machine learning algorithm, hyperparameter tuning is important to train a good multi-label classifier model. The task of selecting the best hyperparameter settings for an algorithm is an optimisation problem. Very limited work has been done on automatic hyperparameter tuning and AutoML in the multi-label domain. This paper attempts to fill this gap by proposing a neural network algorithm, CascadeML, to train multi-label neural network based on cascade neural networks. This method requires minimal or no hyperparameter tuning and also considers pairwise label associations. The cascade algorithm grows the network architecture incrementally in a two phase process as it learns the weights using adaptive first order gradient algorithm, therefore omitting the requirement of preselecting the number of hidden layers, nodes and the learning rate. The method was tested on 10 multi-label datasets and compared with other multi-label classification algorithms. Results show that CascadeML performs very well without hyperparameter tuning.

show abstract

AutoMMLC: An Automated and Multi-objective Method for Multi-label Classification

Del Valle,

Mantovani,

Cerri

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Towards a method for automatically selecting and configuring multi-label classification algorithms

Cited by 19 publications

References 29 publications

Automated Selection and Configuration of Multi-Label Classification Algorithms with Grammar-Based Genetic Programming

Automated Selection and Configuration of Multi-Label Classification Algorithms with Grammar-Based Genetic Programming

CascadeML: An Automatic Neural Network Architecture Evolution and Training Algorithm for Multi-label Classification (Best Technical Paper)

AutoMMLC: An Automated and Multi-objective Method for Multi-label Classification

Contact Info

Product

Resources

About