Rocket Launching: A Universal and Efficient Framework for Training Well-performing Light Net
Explore this paper's citation graph
Summary
This work proposes a universal framework that exploits a booster net to help train the lightweight net for prediction and uses one technique called gradient block to improve the performance of light net and booster net further.
- Type
- article
- Published
- 2017-08-01
- Cited by
- 129
- References
- 34
- Access
- Open access
- OpenAlex
- https://openalex.org/W2748787960
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:3913636
Keywords
Booster (rocketry), Computer science, Net (polyhedron), Inference, Benchmark (surveying)
References
- Deep Canonical Correlation Analysis
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Distilling the Knowledge in a Neural Network
- Going deeper with convolutions
- ImageNet classification with deep convolutional neural networks
- Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
- Signature Verification Using A "Siamese" Time Delay Neural Network
- Deep Residual Learning for Image Recognition
- Model compression
- Reading Digits in Natural Images with Unsupervised Feature Learning
- Learnware: on the future of machine learning
- Wide & Deep Learning for Recommender Systems
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Deep Interest Network for Click-Through Rate Prediction
- ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Do Deep Nets Really Need to be Deep?
- ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
- Densely Connected Convolutional Networks
- Wide Residual Networks
Cited by
- Learning Loss for Knowledge Distillation with Conditional Adversarial Networks
- Low-Resolution Face Recognition in the Wild via Selective Knowledge Distillation
- Knowledge Distillation via Route Constrained Optimization
- Knowledge Squeezed Adversarial Network Compression
- Knowledge Distillation via Instance Relationship Graph
- Privileged Features Distillation for E-Commerce Recommendations
- Knowledge Representing: Efficient, Sparse Representation of Prior Knowledge for Knowledge Distillation
- Deep Interest Evolution Network for Click-Through Rate Prediction
- Improving the Interpretability of Deep Neural Networks with Knowledge Distillation
- Knowledge Distillation with Category-Aware Attention and Discriminant Logit Losses
- Fine-grained Knowledge Fusion for Sequence Labeling Domain Adaptation
- DBP: Discrimination Based Block-Level Pruning for Deep Model Acceleration
- Uncertainty-Aware Multi-Shot Knowledge Distillation for Image-Based Object Re-Identification
- Privileged Features Distillation at Taobao Recommendations
- Knowledge distillation via adaptive instance normalization
- A Survey on Edge Intelligence
- Progressive Learning of Low-Precision Networks for Image Classification
- Knowledge Distillation: A Survey
- Cross-Modal Guidance Network For Sketch-Based 3d Shape Retrieval
- DCAF: A Dynamic Computation Allocation Framework for Online Serving System
Related papers
- AEMB: An Automated Exploit Mitigation Bypassing Solution
- AEG: Automatic Exploit Generation
- Evaluation of Two Host-Based Intrusion Prevention Systems
- Automated Crash Analysis and Exploit Generation with Extendable Exploit Model
- EBF: Event-Based Filter for Exploit Containment
- THE SEARCH ON THE EXPLOIT MACNINATION OF THEME PARK
- On Legal Force of the Exploit Examples of Patents