TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Relay Backpropagation for Effective Learning of Deep Convo...

Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks

Li Shen, Zhouchen Lin, Qingming Huang

2015-12-18Scene ClassificationLong-tail LearningGeneral Classification
PaperPDFCode

Abstract

Learning deeper convolutional neural networks becomes a tendency in recent years. However, many empirical evidences suggest that performance improvement cannot be gained by simply stacking more layers. In this paper, we consider the issue from an information theoretical perspective, and propose a novel method Relay Backpropagation, that encourages the propagation of effective information through the network in training stage. By virtue of the method, we achieved the first place in ILSVRC 2015 Scene Classification Challenge. Extensive experiments on two challenging large scale datasets demonstrate the effectiveness of our method is not restricted to a specific dataset or network architecture. Our models will be available to the research community later.

Results

TaskDatasetMetricValueModel
Image ClassificationCOCO-MLTAverage mAP46.97RS(ResNet-50)
Image ClassificationVOC-MLTAverage mAP75.38RS(ResNet-50)
Few-Shot Image ClassificationCOCO-MLTAverage mAP46.97RS(ResNet-50)
Few-Shot Image ClassificationVOC-MLTAverage mAP75.38RS(ResNet-50)
Generalized Few-Shot ClassificationCOCO-MLTAverage mAP46.97RS(ResNet-50)
Generalized Few-Shot ClassificationVOC-MLTAverage mAP75.38RS(ResNet-50)
Long-tail LearningCOCO-MLTAverage mAP46.97RS(ResNet-50)
Long-tail LearningVOC-MLTAverage mAP75.38RS(ResNet-50)
Generalized Few-Shot LearningCOCO-MLTAverage mAP46.97RS(ResNet-50)
Generalized Few-Shot LearningVOC-MLTAverage mAP75.38RS(ResNet-50)

Related Papers

Towards Scalable and Generalizable Earth Observation Data Mining via Foundation Model Composition2025-06-25Mitigating Spurious Correlations with Causal Logit Perturbation2025-05-21A Challenge to Build Neuro-Symbolic Video Agents2025-05-20EarthSynth: Generating Informative Earth Observation with Diffusion Models2025-05-17Energy efficiency analysis of Spiking Neural Networks for space applications2025-05-16Minimizing Risk Through Minimizing Model-Data Interaction: A Protocol For Relying on Proxy Tasks When Designing Child Sexual Abuse Imagery Detection Models2025-05-10Low-Complexity Acoustic Scene Classification with Device Information in the DCASE 2025 Challenge2025-05-03FrogDogNet: Fourier frequency Retained visual prompt Output Guidance for Domain Generalization of CLIP in Remote Sensing2025-04-23