Application of Smoothing Labels to Alleviate Overconfident of the GAN's Discriminator
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
A Deep Convolutional Generative Adversarial Network (DCGAN) suffers from a vanishing gradient issue in the generator due to the overconfidence of the discriminator.This paper explores the effects of using noise injection and gradually changing label smoothing (CLS) towards hard labels and two-sided label smoothing to enhance the stability of the DCGAN.Different models are trained on CIFAR-10 datasets that contains 60,000 3232 color images divided into 10 categories and CIFAR-100 datasets that contains 60,000 3232 color images divided into 100 categories, compared with each other using Fr chet Inception distance (FID), and Inception Score (IS) evaluation metrics.A noticeable improvement in generalization was found in almost all cases, and the best was when using CLS for both real and fake labels of two-sided smoothing labels.The modified DCGAN performs better than traditional DCGAN, boosting the best Fr chet Inception distance from 132.31 to 95.52 and the Inception Score (IS) from 25.123 to 64.27 on the CIFAR-10 dataset, the FID from 137.84 to 109.42, and the IS from 19.65 to 61.04 on the challenging CIFAR-100 dataset.
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.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 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 it