# CascadeML: An Automatic Neural Network Architecture Evolution and Training Algorithm for Multi-label Classification

## TL;DR

CascadeML introduces an automatic neural network architecture evolution method for multi-label classification that minimizes hyperparameter tuning and leverages label associations, achieving strong performance across multiple datasets.

## Contribution

It presents a novel cascade neural network algorithm that automatically evolves architecture and considers label correlations, reducing the need for hyperparameter tuning in multi-label classification.

## Key findings

- CascadeML performs well without hyperparameter tuning.
- It outperforms other algorithms on 10 multi-label datasets.
- The method effectively considers label associations during training.

## Abstract

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.