Tricks and Plugins to GBM on Images and Sequences

TL;DR

This paper introduces new boosting algorithms for CNNs and transformers that simplify architecture search, improve performance, and reduce training time on image and sequence tasks.

Contribution

It proposes a novel boosting framework with subgrid selection and importance sampling for deep networks, enhancing efficiency and accuracy.

Findings

Outperforms benchmarks on fine-grained classification tasks

Reduces manual effort in architecture tuning

Achieves superior performance with lower training time

Abstract

Convolutional neural networks (CNNs) and transformers, which are composed of multiple processing layers and blocks to learn the representations of data with multiple abstract levels, are the most successful machine learning models in recent years. However, millions of parameters and many blocks make them difficult to be trained, and sometimes several days or weeks are required to find an ideal architecture or tune the parameters. Within this paper, we propose a new algorithm for boosting Deep Convolutional Neural Networks (BoostCNN) to combine the merits of dynamic feature selection and BoostCNN, and another new family of algorithms combining boosting and transformers. To learn these new models, we introduce subgrid selection and importance sampling strategies and propose a set of algorithms to incorporate boosting weights into a deep learning architecture based on a least squares objective function. These algorithms not only reduce the required manual effort for finding an appropriate network architecture but also result in superior performance and lower running time. Experiments show that the proposed methods outperform benchmarks on several fine-grained classification tasks.