Optimization of Deep Neural Networks for Enhanced Efficiency in Small Scale Autonomous Vehicles
Bibliographic record
Abstract
Autonomous vehicles of the contemporary era constitute a sophisticated blend of artificial intelligence and electronic components.These vehicles operate autonomously by employing neural networks trained to interpret visual input from multiple onboard cameras and subsequently produce corresponding steering angles.However, the existing neural networks are characterized by their substantial scale, necessitating substantial GPU resources, and are prone to latency issues and complex architectural requirements.These factors render these networks unsuitable for small-scale applications where latency, complex architecture, and expensive hardware are prohibitive.This paper proposes a methodology for optimizing these neural networks for small-scale operations while preserving their accuracy and precision.This is achieved through a fine-tuning process that customizes the architecture and modifies various functional values and their parameters, resulting in a deep neural network tailored for small-scale applications.This optimized network boasts a simpler architecture, lower storage requirements, and reduced demand for GPU resources.The network is developed, trained, and evaluated using TensorFlow, a widely employed API for machine learning applications.The optimized network offers several advantages, including reduced latency, a customizable architecture, minimized memory requirements, and decreased GPU demand, making it a viable solution for various applications.The paper provides a detailed exploration of the development of this bespoke deep neural network and its potential implications for the future of small-scale autonomous vehicles.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".