How Does a Neural Network's Architecture Impact Its Robustness to Noisy\n Labels?
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
Noisy labels are inevitable in large real-world datasets. In this work, we\nexplore an area understudied by previous works -- how the network's\narchitecture impacts its robustness to noisy labels. We provide a formal\nframework connecting the robustness of a network to the alignments between its\narchitecture and target/noise functions. Our framework measures a network's\nrobustness via the predictive power in its representations -- the test\nperformance of a linear model trained on the learned representations using a\nsmall set of clean labels. We hypothesize that a network is more robust to\nnoisy labels if its architecture is more aligned with the target function than\nthe noise. To support our hypothesis, we provide both theoretical and empirical\nevidence across various neural network architectures and different domains. We\nalso find that when the network is well-aligned with the target function, its\npredictive power in representations could improve upon state-of-the-art (SOTA)\nnoisy-label-training methods in terms of test accuracy and even outperform\nsophisticated methods that use clean labels.\n
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 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.000 | 0.001 |
| 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