NeurIPS’22 Cross-Domain MetaDL Challenge:Results and lessons learned
Bibliographic record
Abstract
Deep neural networks have demonstrated the ability to outperform humans in multiple tasks, but they often require substantial amounts of data and computational resources. These resources may be limited in certain fields. Meta-learning seeks to overcome these challenges by utilizing past task experiences to efficiently solve new tasks, achieving better performance with limited training data and modest computational resources. To further advance the ChaLearn MetaDL competition series, we organized the Cross-Domain MetaDL Challenge for NeurIPS’22. This challenge aimed to solve “any-way” and “any-shot” tasks from 10 domains through cross-domain meta-learning. In this paper, authored collaboratively by the competition organizers, top-ranked participants, and external collaborators, we describe the technical aspects of the competition, baseline methods, and top-ranked approaches that have been open-sourced. Additionally, we provide a detailed analysis of the competition results. Lessons learned from this competition include the critical role of pre-trained backbones, the necessity of preventing overfitting, and the significance of using data augmentation or domain adaptation techniques in conjunction with extra optimizations to improve performance.
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 | 0.017 | 0.021 |
| Meta-epidemiology (narrow) | 0.007 | 0.001 |
| Meta-epidemiology (broad) | 0.004 | 0.003 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.005 | 0.008 |
| Research integrity | 0.006 | 0.006 |
| Insufficient payload (model declined to judge) | 0.013 | 0.015 |
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".