Using deep learning approaches to overcome limited dataset issues within semiconductor domain
Bibliographic record
Abstract
Today, in semiconductor manufacturing, wafer failures are frequent problems with the production lines. To increase the production yield, images are the most important pieces of data used to detect defect-free wafers. However, there are many tools that can be installed specifically to monitor production lines, inspect mapped defects and detect the main causes of die failures by using wafers images during semiconductor manufacturing process. The underlying objective is to overcome the need for a physical check on the wafer which are in most cases too long. Thus, the need to have a design that will measure and detect these visual faults in an automated fashion is a big challenge for the industry. Recently, deep learning approaches have proven to be a suitable way to overcome this issue. However, they rely on the availability of sufficiently representative datasets which is not our case: data on anomalies is scarce. The goal of this paper is to evaluate state of the art deep learning methodology such as GAN and VAE to overcome this challenge. Implementation results show that the GAN architecture achieves a convincing image generation in a limited sample setting, while the VAE architecture fails to converge at training time.
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".