# Sparsely ensembled convolutional neural network classifiers via reinforcement learning

arXiv: https://arxiv.org/abs/2102.03921 · HTML: https://mlsanity.com/paper/2102.03921/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 29 references that are in the map.
- First posted on arXiv 7 Feb 2021.

## Builds on (most cited first, 13 of 13)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Adaptive Computation Time for Recurrent Neural Networks](https://mlsanity.com/paper/1603.08983/) (2016)
- [Recurrent Models of Visual Attention](https://mlsanity.com/paper/1406.6247/) (2014)
- [Spatially Adaptive Computation Time for Residual Networks](https://mlsanity.com/paper/1612.02297/) (2016)
- [Multi-label Image Recognition by Recurrently Discovering Attentional Regions](https://mlsanity.com/paper/1711.02816/) (2017)
- [Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning](https://mlsanity.com/paper/1804.03312/) (2018)
- [Path-Restore: Learning Network Path Selection for Image Restoration](https://mlsanity.com/paper/1904.10343/) (2019)
- [Deciding How to Decide: Dynamic Routing in Artificial Neural Networks](https://mlsanity.com/paper/1703.06217/) (2017)
- [Dynamic Computational Time for Visual Attention](https://mlsanity.com/paper/1703.10332/) (2017)
- [Learning Policies for Adaptive Tracking with Deep Feature Cascades](https://mlsanity.com/paper/1708.02973/) (2017)
- [Hierarchical Object Detection with Deep Reinforcement Learning](https://mlsanity.com/paper/1611.03718/) (2016)
- [Deep Incremental Boosting](https://mlsanity.com/paper/1708.03704/) (2017)

## Built on by (most cited first, 2 of 2)

- [Forging Vision Foundation Models for Autonomous Driving: Challenges, Methodologies, and Opportunities](https://mlsanity.com/paper/2401.08045/) (2024)
- [Neural network task specialization via domain constraining](https://mlsanity.com/paper/2504.19592/) (2025)

## Abstract (arXiv)

We consider convolutional neural network (CNN) ensemble learning with the objective function inspired by least action principle; it includes resource consumption component. We teach an agent to perceive images through the set of pre-trained classifiers and want the resulting dynamically configured system to unfold the computational graph with the trajectory that refers to the minimal number of operations and maximal expected accuracy. The proposed agent's architecture implicitly approximates the required classifier selection function with the help of reinforcement learning. Our experimental results prove, that if the agent exploits the dynamic (and context-dependent) structure of computations, it outperforms conventional ensemble learning.