paper-with-me

Papers

Neural Network Compression Using Higher-Order Statistics and AuxiliaryReconstruction Losses

2020-06-15 · Christos Chatzikonstantinou, Georgios Th. Papadopoulos, Kosmas Dimitropoulos, Petros Daras

In this paper, the problem of pruning and compressingthe weights of various layers of deep neural networks is in-vestigated. The proposed method aims to remove redundantfilters from the network to reduce computational complex-ity and storage requirements, while improving the perfor-mance of the original network. More specifically, a novelfilter selection criterion is introduced based on the fact thatfilters whose weights follow a Gaussian distribution corre-spond to hidden units that do not capture important aspectsof data. To this end, Higher Order Statistics (HOS) areused and filters with low cumulant values that do not de-viate significantly from Gaussian distribution are identifiedand removed from the network. In addition, a novel prun-ing strategy is proposed aiming to decide on the pruningratio of each layer using the Shapiro-Wilk normality test.The use of auxiliary MSE losses (intermediate and afterthe softmax layer) during the fine-tuning phase further im-proves the overall performance of the compressed network.Extensive experiments with different network architecturesand comparison with state-of-the-art approaches on well-known public datasets, such as CIFAR-10, CIFAR-100 andILSCVR-12, demonstrate the great potential of the proposedapproach.

📄 PDF Abstract BibTeX

Code (1)

chatzikon/DNN-COMPRESSION pytorch

Tasks

Neural Network Compression

Methods 이 논문이 사용한 방법론

Pruning 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

ELiC: Efficient LiDAR Geometry Compression via Cross-Bit-depth Feature Propagation and Bag-of-Encoders

2025-11-18 · Junsik Kim, Gun Bang, Soowoong Kim arxiv

Hierarchical LiDAR geometry compression encodes voxel occupancies from low to high bit-depths, yet prior methods treat each depth independently and re-estimate local context from coordinates at every level, limiting comp…

A Bayesian Optimization Framework for Neural Network Compression

2019-10-01 · ICCV 2019 10 · Xingchen Ma, Amal Rannen Triki, Maxim Berman, Christos Sagonas 외

Neural network compression is an important step for deploying neural networks where speed is of high importance, or on devices with limited memory. It is necessary to tune compression parameters in order to achieve the d…

Bayesian OptimizationKnowledge DistillationNeural Network Compression

On computing and the complexity of computing higher-order $U$-statistics, exactly

2025-08-18 · Xingyu Chen, Ruiqi Zhang, Lin Liu arxiv

Higher-order $U$-statistics abound in fields such as statistics, machine learning, and computer science, but are known to be highly time-consuming to compute in practice. Despite their widespread appearance, a comprehens…

High Order Structure Descriptors for Scene Images

2014-10-15 · Wenya Zhu, Xiankai Lu, Tao Xu, Ziyi Zhao

Structure information is ubiquitous in natural scene images and it plays an important role in scene representation. In this paper, third order structure statistics (TOSS) and fourth order structure statistics (FOSS) are …

Scene ClassificationVocal Bursts Intensity Prediction

A theory of learning data statistics in diffusion models, from easy to hard

2026-03-13 · Lorenzo Bardone, Claudia Merger, Sebastian Goldt arxiv

While diffusion models have emerged as a powerful class of generative models, their learning dynamics remain poorly understood. We address this issue first by empirically showing that standard diffusion models trained on…