Sensitivity of a Convolutional Neural Network for Different Pooling Layers in Spatial Domain Steganalysis
Bibliographic record
Abstract
In the modern era, numerous research studies consistently affirm the superior performance of Convolutional Neural Networks (CNNs) over traditional machine learning methods in steganalysis, a technique used to detect hidden data through steganography.Deep Learning (DL), particularly CNNs, is a powerful tool for steganalysis because it can handle large datasets effectively.Despite CNNs being widely used in various research areas, previous steganalysis studies have primarily focused on improving image classification (cover or stego), often neglecting a thorough exploration of the experimental setup.This research aims to assess the sensitivity of a CNN-based steganalysis model by investigating the impact of different pooling layers on state-of-the-art models.The experiments involve five recently proposed models.Significantly, the choice of pooling layers goes beyond mere classification improvement; it also addresses overfitting.The experimental results reveal significant diversity based on the selected pooling layers, namely the maximum, average, and mixed pooling, emphasizing the importance of optimizing objectives when choosing a particular pooling approach.This highlights the evolving nature of this field of study and the need for careful consideration in pooling layer selection for effective steganalysis.
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 | 0.004 | 0.017 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".