Loss-Aware Feature-Map Pruning in Convolutional Neural Networks Using Multi-Armed Bandits
Summary
This paper introduces a loss-aware feature-map pruning framework for convolutional neural networks using multi-armed bandits (UCB1 and Thompson Sampling) to selectively remove redundant feature maps, reducing computational cost while preserving accuracy.
View Cached Full Text
Cached at: 07/28/26, 06:24 AM
# Loss-Aware Feature-Map Pruning in Convolutional Neural Networks Using Multi-Armed Bandits Source: [https://arxiv.org/abs/2607.22564](https://arxiv.org/abs/2607.22564) [View PDF](https://arxiv.org/pdf/2607.22564) > Abstract:Convolutional neural networks often contain redundant feature maps that increase storage and inference cost\. This paper presents a loss\-aware feature\-map pruning framework using multi\-armed bandits\. Feature\-map pruning is structured because it removes complete convolutional output channels and their producing filters rather than isolated scalar weights\. Each candidate feature map is treated as an arm\. At each play time, one map is temporarily masked and evaluated on a sampled mini\-batch; the map is then restored and the observed loss change is converted into a safe\-removal reward\. After a fixed play budget, candidate maps are ranked by learned scores and the top\-k maps are permanently removed with their filters, biases and corresponding next\-layer input\-channel kernels\. The study evaluates UCB1 and Thompson Sampling, compares them with direct/oracle\-style evaluation on LeNet/MNIST, and extends the evaluation to MNIST, CIFAR\-10, CIFAR\-100, SVHN, CUB\-200\-2011 and Oxford Flowers 102\. Results show that UCB1 and Thompson Sampling preserve accuracy close to unpruned models while removing feature maps and reducing convolutional computation\. Friedman and Nemenyi tests show that UCB1 obtains the highest mean rank, followed by Thompson Sampling; both significantly outperform greedy and magnitude\-based pruning while remaining statistically comparable to the original unpruned model\. ## Submission history From: Salem Ameen \[[view email](https://arxiv.org/show-email/4f89522d/2607.22564)\] **\[v1\]**Fri, 29 May 2026 20:03:08 UTC \(560 KB\)
Similar Articles
Structured Neuron Pruning in Deep Neural Networks Using Multi-Armed Bandits
This paper proposes a novel structured neuron pruning framework for deep neural networks using multi-armed bandit algorithms, demonstrating effectiveness on various tasks.
Pruning Binarized Neural Networks: A Dedicated Framework and Globally Weighted Algorithms
This paper introduces a PyTorch-based framework for pruning binarized neural networks and proposes a novel global weighting pruning method that achieves superior accuracy and pruning rates, enabling efficient deployment on edge hardware like FPGAs.
An AI4AI Framework for Visual Token Pruning
AutoPrune is a training-free framework that uses LLMs to automatically design visual-token pruning policies for multimodal LLMs via a domain-specific language and residual search, achieving high efficiency with minimal performance loss (99% performance retained while removing 94.4% of visual tokens).
Pruning Deep Neural Networks via the Marchenko--Pastur Distribution
This paper presents a Marchenko-Pastur random matrix approach to pruning deep neural networks, offering theoretical guarantees and achieving strong accuracy retention with minimal fine-tuning on ImageNet for ViT and CNN architectures.
Unifying Depth and Width Pruning for LLMs via Binary Knapsack Optimization
This paper introduces Sniper, a two-stage structured pruning framework for LLMs that uses binary knapsack optimization to unify depth and width pruning, achieving near-exact compression ratio adherence and improved performance retention across multiple architectures.