# Oracle Supervision Transfers for Hyperparameter Prediction in Model-Based Image Denoising

arXiv: https://arxiv.org/abs/2605.20479 · HTML: https://mlsanity.com/paper/2605.20479/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 29 references that are in the map.
- First posted on arXiv 19 May 2026.

## Builds on (most cited first, 15 of 15)

- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [HyperNetworks](https://mlsanity.com/paper/1609.09106/) (2016)
- [On First-Order Meta-Learning Algorithms](https://mlsanity.com/paper/1803.02999/) (2018)
- [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://mlsanity.com/paper/2301.00808/) (2023)
- [Denoising Diffusion Models for Plug-and-Play Image Restoration](https://mlsanity.com/paper/2305.08995/) (2023)
- [Dynamic Filter Networks](https://mlsanity.com/paper/1605.09673/) (2016)
- [Towards Learning Universal Hyperparameter Optimizers with Transformers](https://mlsanity.com/paper/2205.13320/) (2022)
- [Few-Shot Bayesian Optimization with Deep Kernel Surrogates](https://mlsanity.com/paper/2101.07667/) (2021)
- [Tuning-free Plug-and-Play Proximal Algorithm for Inverse Imaging Problems](https://mlsanity.com/paper/2002.09611/) (2020)
- [Practical Transfer Learning for Bayesian Optimization](https://mlsanity.com/paper/1802.02219/) (2018)
- [Learning Regularization Parameters of Inverse Problems via Deep Neural Networks](https://mlsanity.com/paper/2104.06594/) (2021)
- [Bilevel parameter learning for higher-order total variation regularisation models](https://mlsanity.com/paper/1508.07243/) (2015)
- [Computing Multiple Image Reconstructions with a Single Hypernetwork](https://mlsanity.com/paper/2202.11009/) (2022)
- [Deep unrolling for learning optimal spatially varying regularisation parameters for Total Generalised Variation](https://mlsanity.com/paper/2502.16532/) (2025)

## Abstract (arXiv)

Hyperparameter prediction is a critical practical bottleneck for model-based image denoisers, ranging from classical TV/TGV variational solvers to modern diffusion-based models such as DiffPIR. While existing learned predictors can achieve near-oracle performance, this approach scales poorly: each new configuration conventionally requires its own oracle-labeled training set, and each label requires a hierarchical grid search evaluated against clean ground truth. We therefore ask whether oracle supervision collected on source configurations can transfer to target configurations with few or no target oracle labels. We propose HyperDn, a single configuration-conditioned predictor that pools oracle supervision across source configurations and predicts heterogeneous hyperparameters for new denoiser--noise configurations. In a cross-paradigm experiment, HyperDn transfers from relatively cheap TV/TGV variational sources to more expensive diffusion-based DiffPIR. With only 2 target oracle labels, it reaches 30.23,dB, within 0.90,dB of the oracle, and outperforms the 64-label per-configuration predictor trained from scratch, using 1/32 as many target labels as that baseline point. Without any target oracle labels, HyperDn also reaches near-oracle PSNR on two unseen mixtures of seen noise types and on transfer from relatively cheap 96× 96 source images to 512× 768 targets. Together, these results show that expensive oracle supervision for hyperparameter prediction can be transferred from source to new target configurations, reducing the need to rebuild oracle labels for each new denoising configuration.