Flexi-Compression: A Flexible Model Compression Method for Autonomous Driving
Bibliographic record
Abstract
Benefiting from the rapid development of convolutional neural networks, computer vision-based autonomous driving technologies are gradually being deployed in vehicles. However, these neural networks typically have a large number of parameters and extremely high computational cost, making them difficult to deploy in autonomous vehicles with limited storage and computational power. In this thesis, we propose an innovative model compression approach to compress convolutional neural networks in autonomous driving algorithms, which we call Flexi-Compression. Flexi-Compression first modifies the model structure by replacing the traditional convolutional layers with our proposed Flexi-CP(flexible compression) module, thus reducing the computation of the convolutional layers. Then, we leverage knowledge distillation to enable the compressed model to quickly acquire the knowledge of the original model. In addition, we use a Flexi-Batch Normalization layer to prune the model and finally further reduce the model size by model quantization. We compress the VGG-16 network(Visual Geometry Group Network) using our proposed model compression algorithm, which is a commonly used backbone network in autonomous driving algorithms. On the CIFAR-10 dataset, our compression method can reduce the parameters of the VGG-16 network by 86% and the computation by 87% with 4% loss of accuracy. To verify the effectiveness of our compression algorithm in real-world applications, we also compress an autonomous driving algorithm and achieve excellent performance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".