Joint Progressive Knowledge Distillation and Unsupervised Domain\n Adaptation
Bibliographic record
Abstract
Currently, the divergence in distributions of design and operational data,\nand large computational complexity are limiting factors in the adoption of CNNs\nin real-world applications. For instance, person re-identification systems\ntypically rely on a distributed set of cameras, where each camera has different\ncapture conditions. This can translate to a considerable shift between source\n(e.g. lab setting) and target (e.g. operational camera) domains. Given the cost\nof annotating image data captured for fine-tuning in each target domain,\nunsupervised domain adaptation (UDA) has become a popular approach to adapt\nCNNs. Moreover, state-of-the-art deep learning models that provide a high level\nof accuracy often rely on architectures that are too complex for real-time\napplications. Although several compression and UDA approaches have recently\nbeen proposed to overcome these limitations, they do not allow optimizing a CNN\nto simultaneously address both. In this paper, we propose an unexplored\ndirection -- the joint optimization of CNNs to provide a compressed model that\nis adapted to perform well for a given target domain. In particular, the\nproposed approach performs unsupervised knowledge distillation (KD) from a\ncomplex teacher model to a compact student model, by leveraging both source and\ntarget data. It also improves upon existing UDA techniques by progressively\nteaching the student about domain-invariant features, instead of directly\nadapting a compact model on target domain data. Our method is compared against\nstate-of-the-art compression and UDA techniques, using two popular\nclassification datasets for UDA -- Office31 and ImageClef-DA. In both datasets,\nresults indicate that our method can achieve the highest level of accuracy\nwhile requiring a comparable or lower time complexity.\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.
How this classification was reachedexpand
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 | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".