Self-Tuning Networks: Bilevel Optimization of Hyperparameters using\n Structured Best-Response Functions
Bibliographic record
Abstract
Hyperparameter optimization can be formulated as a bilevel optimization\nproblem, where the optimal parameters on the training set depend on the\nhyperparameters. We aim to adapt regularization hyperparameters for neural\nnetworks by fitting compact approximations to the best-response function, which\nmaps hyperparameters to optimal weights and biases. We show how to construct\nscalable best-response approximations for neural networks by modeling the\nbest-response as a single network whose hidden units are gated conditionally on\nthe regularizer. We justify this approximation by showing the exact\nbest-response for a shallow linear network with L2-regularized Jacobian can be\nrepresented by a similar gating mechanism. We fit this model using a\ngradient-based hyperparameter optimization algorithm which alternates between\napproximating the best-response around the current hyperparameters and\noptimizing the hyperparameters using the approximate best-response function.\nUnlike other gradient-based approaches, we do not require differentiating the\ntraining loss with respect to the hyperparameters, allowing us to tune discrete\nhyperparameters, data augmentation hyperparameters, and dropout probabilities.\nBecause the hyperparameters are adapted online, our approach discovers\nhyperparameter schedules that can outperform fixed hyperparameter values.\nEmpirically, our approach outperforms competing hyperparameter optimization\nmethods on large-scale deep learning problems. We call our networks, which\nupdate their own hyperparameters online during training, Self-Tuning Networks\n(STNs).\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 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 | 0.009 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 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".